Prompt-driven scene synthesis allows users to generate complete 3D environments from textual descriptions. Current text-to-scene methods often struggle with complex geometries and object transformations, and tend to show weak adherence to complex instructions. We address these limitations by introducing Reason-3D, a text-to-scene model powered by large reasoning models (LRMs). Reason-3D integrates object retrieval using captions covering physical, functional, and contextual attributes. Reason-3D then places the selected objects based on implicit and explicit layout constraints, and refines their positions with collision-aware spatial reasoning. Evaluated on instructions ranging from simple to complex indoor configurations, Reason-3D significantly outperforms previous methods in human-rated visual fidelity, adherence to constraints, and asset retrieval quality. Beyond its contribution to the field of text-to-scene generation, our work showcases the advanced spatial reasoning abilities of modern LRMs. Additionally, we release the codebase to further the research in object retrieval and placement with LRMs.

Speculative decoding accelerates language model inference by separating generation into fast drafting and parallel verification. Its main limitation is drafter–verifier misalignment, which limits token acceptance and reduces overall effectiveness. While small drafting heads trained from scratch compensate with speed, they struggle when verification dominates latency or when inputs are out of distribution. In contrast, pretrained drafters, though slower, achieve higher acceptance rates thanks to stronger standalone generation capabilities, making them competitive when drafting latency is negligible relative to verification or communication overhead. In this work, we aim to improve the acceptance rates of pretrained drafters by introducing a lightweight dynamic alignment mechanism: a steering vector computed from the verifier’s hidden states and injected into the pretrained drafter. Compared to existing offline alignment methods such as distillation, our approach boosts the number of accepted tokens by up to 35% under standard sampling and 22% under greedy sampling, all while incurring negligible computational overhead. Importantly, our approach can be retrofitted to existing architectures and pretrained models, enabling rapid adoption.

Positional Encodings (PEs) are essential for injecting structural information into Graph Neural Networks (GNNs), particularly Graph Transformers, yet their empirical impact remains insufficiently understood. We introduce a unified benchmarking framework that decouples PEs from architectural choices, enabling a fair comparison across 8 GNN and Transformer models, 9 PEs, and 10 synthetic and real-world datasets. Across more than 500 model-PE-dataset configurations, we find that commonly used expressiveness proxies, including Weisfeiler-Lehman distinguishability, do not reliably predict downstream performance. In particular, highly expressive PEs frequently fail to improve, and can even degrade performance on real-world tasks. At the same time, we identify several simple and previously overlooked model-PE combinations that match or outperform recent state-of-the-art methods. Our results demonstrate the strong task-dependence of PEs and underscore the need for empirical validation beyond theoretical expressiveness. To support reproducible research, we release an open-source benchmarking framework for evaluating PEs for graph learning tasks.

Achieving agreement among distributed parties is a fundamental task in modern systems, underpinning applications such as consensus in blockchains, coordination in cloud infrastructure, and fault tolerance in critical services. However, this task can be intensive, often requiring a large number of messages to be exchanged as well as many rounds of communication, especially in the presence of Byzantine faults. This makes efficiency a central challenge in the design of practical agreement protocols.

In the online Min-cost Perfect Matching with Delays (\(\textsf{MPMD}\)) problem, \(m\) requests in a metric space are submitted at different times by an adversary. The goal is to match all requests while (i) minimizing the sum of the distances between matched pairs as well as (ii) how long each request remained unmatched after it appeared.

Contract clause retrieval is foundational to contract drafting because lawyers rarely draft contracts from scratch; instead, they locate and revise the most relevant precedent clauses. We introduce the Atticus Clause Retrieval Dataset (ACORD), the first expert-annotated benchmark specifically designed for contract clause retrieval to support contract drafting tasks. ACORD focuses on complex contract clauses such as Limitation of Liability, Indemnification, Change of Control, and Most Favored Nation. It includes 114 queries and over 126,000 query-clause pairs, each ranked on a scale from 1 to 5 stars. The task is to find the most relevant precedent clauses to a query. The bi-encoder retriever paired with pointwise LLMs re-rankers shows promising results. However, substantial improvements are still needed to manage the complex legal work typically undertaken by lawyers effectively. As the first expert-annotated benchmark for contract clause retrieval, ACORD can serve as a valuable IR benchmark for the NLP community.

Given the low throughput of blockchains like Bitcoin and Ethereum, scalability - the ability to process an increasing number of transactions - has become a central focus of blockchain research. One promising approach is the parallelization of transaction execution across multiple threads. However, achieving efficient parallelization requires a redesign of the incentive structure within the fee market. Currently, the fee market does not differentiate between transactions that access multiple high-demand storage keys (i.e., unique identifiers for individual data entries) versus a single low-demand one, as long as they require the same computational effort. Addressing this discrepancy is crucial for enabling more effective parallel execution. In this work, we aim to bridge the gap between the current fee market and the need for parallel execution by exploring alternative fee market designs. To this end, we propose a framework consisting of two key components: a Gas Computation Mechanism (GCM), which quantifies the load a transaction places on the network in terms of parallelization and computation, measured in units of gas, and a Transaction Fee Mechanism (TFM), which assigns a price to each unit of gas. We additionally introduce a set of desirable properties for a GCM, propose several candidate mechanisms, and evaluate them against these criteria. Our analysis highlights two strong candidates: the weighted area GCM, which integrates smoothly with existing TFMs such as EIP‑1559 and satisfies a broad subset of the outlined properties, and the time-proportional makespan GCM, which assigns gas costs based on the context of the entire block’s schedule and, through this dependence on the overall execution outcome, captures the dynamics of parallel execution more accurately.

Layer 2 rollups are rapidly absorbing DeFi activity, securing over $40 billion and accounting for nearly half of Ethereum’s DEX volume by Q1 2025, yet their MEV dynamics remain understudied. We address this gap by defining and quantifying optimistic MEV, a form of speculative, on-chain MEV whose detection and execution logic reside largely on-chain in smart contracts. As a result of their speculative nature and lack of off-chain opportunity verification, optimistic MEV transactions frequently decide not to execute any trades. In this work, we focus on cyclic arbitrage, which we find is predominantly executed as optimistic MEV on Layer 2s. Using our multi-stage identification pipeline on Arbitrum, Base, and Optimism, we show that in Q1 2025, transactions from cyclic arbitrage contracts account for over 50% of on-chain gas on Base and Optimism and 7% on Arbitrum, driven mainly by "interaction" probes (on-chain computations searching for arbitrage). This speculative probing indicates that cyclic arbitrage on Layer 2s is predominantly executed as optimistic MEV and contributes to generally keeping blocks on Base and Optimism persistently full. Despite consuming over half of on-chain gas, these optimistic MEV transactions pay less than one quarter of total gas fees. Cross-network comparison reveals divergent success rates, differing patterns of code reuse, and sensitivity to varying sequencer ordering and block production times. Finally, OLS regressions link optimistic MEV trade count to ETH volatility, retail trading activity, and DEX aggregator usage. Together, these findings show that optimistic MEV has become a major source of persistent spam-like transaction activity on Layer 2s, dominating blockspace with low-value probes and reshaping the composition of on-chain activity.

In lossy image compression, models face the challenge of hallucinating details or generating out-of-distribution samples due to the information bottleneck. Sometimes, hallucinations are necessary to generate in-distribution samples, but the optimal level depends on image content, as humans notice small changes that affect meaning. We propose ConHa, a compression method that dynamically balances hallucination levels based on content. We train a model to predict user preferences on detail and hallucination levels and use this prediction to adjust the perceptual weight in the reconstruction loss. To evaluate our modes performance, we gathered 1,531 comparisons from 40 participants across three bitrates, see Figure 1. Our model outperforms both Hyperprior [1] and HiFiC [2], achieving a middle ground. The performance improvement over HiFiC holds true at all bitrates. When the perceptual loss weight is fixed to the dataset mean, and not predicted for each image, performance drops to near HiFiC levels. This shows the importance of conditioning the perceptual loss weight on the image for optimal performance. ConHa selectively hallucinates details based on content, improving realism and outperforming all baseline models.

Recent advancements have brought generated music closer to human-created compositions, yet evaluating these models remains challenging. While human preference is the gold standard for assessing quality, translating these subjective judgments into objective metrics, particularly for text-audio alignment and music quality, has proven difficult. In this work, we generate 6k songs using 12 state-of-the-art models and conduct a survey of 15k pairwise audio comparisons with 2.5k human participants to evaluate the correlation between human preferences and widely used metrics. To the best of our knowledge, this work is the first to rank current state-of-the-art music generation models and metrics based on human preference. To further the field of subjective metric evaluation, we provide open access to our dataset of generated music and human evaluations.

Recent multimodal methods for lyrics alignment have relied on large datasets. Our approach introduces a box loss that directly incorporates timestamp information into the loss function, enabling precise alignment and competitive results even with limited training data. We also address the noise present in the public DALI dataset, conducting a thorough cleaning process to improve the quality of training data. Finally, we propose JamendoLyrics++, a substantial extension of the common JamendoLyrics evaluation dataset, offering improved genre diversity for better evaluation of lyrics alignment systems.

While neural vocoders have made significant progress in high-fidelity speech synthesis, their application on polyphonic music has remained underexplored. In this work, we propose DisCoder, a neural vocoder that leverages a generative adversarial encoder-decoder architecture informed by a neural audio codec to reconstruct high-fidelity 44.1 kHz audio from mel spectrograms. Our approach first transforms the mel spectrogram into a lower-dimensional representation aligned with the Descript Audio Codec (DAC) latent space before reconstructing it to an audio signal using a fine-tuned DAC decoder. DisCoder achieves state-of-the-art performance in music synthesis on several objective metrics and in a MUSHRA listening study. Our approach also shows competitive performance in speech synthesis, highlighting its potential as a universal vocoder.

We introduce DiscoDiff, a text-to-music generative model that utilizes two latent diffusion models to produce high-fidelity 44.1kHz music hierarchically. Our approach significantly enhances audio quality through a coarse-to-fine generation strategy, leveraging residual vector quantization from the Descript Audio Codec. We consolidate this coarse-to-fine design through an important observation that the audio latent representation can be split into a primary and secondary part, controlling music content and details accordingly. We validate the effectiveness of our approach and text-audio alignment through various objective metrics. Furthermore, we provide access to high-quality synthetic captions for the MTG-Jamendo and FMA datasets, as well as open-sourcing DiscoDiff’s codebase and model checkpoints.

We introduce BLAP, a model capable of generating high-quality captions for music. BLAP leverages a fine-tuned CLAP audio encoder and a pre-trained Flan-T5 large language model. To achieve effective cross-modal alignment between music and language, BLAP utilizes a Querying Transformer, allowing us to obtain state-of-the-art performance using 6x less data compared to previous models. This is a critical consideration given the scarcity of descriptive music data and the subjective nature of music interpretation. We provide qualitative examples demonstrating BLAP’s ability to produce realistic captions for music, and perform a quantitative evaluation on three datasets. BLAP achieves a relative improvement on FENSE compared to previous models of 3.5%, 6.5%, and 7.5% on the MusicCaps, Song Describer, and YouTube8m-MTC datasets, respectively. The codebase is available at https://github.com/ETH-DISCO/blap.

In this work, we introduce AutoSing, a novel framework designed to generate diverse and high-quality singing voices from provided lyrics and musical accompaniment. AutoSing extends an existing semantic token-based text-to-speech approach by incorporating musical accompaniment as an additional conditioning input. This enables AutoSing to synchronize its vocal output with the rhythm and melodic nuances of the accompaniment while adhering to the provided lyrics. Our contributions include a novel training scheme for autoregressive audio models applied to singing voice synthesis, as well as ablation studies to identify the best way to condition generation on musical accompaniment. We measure AutoSing’s performance with subjective listening tests, demonstrating its capability to generate coherent and creative singing voices. Furthermore, we open-source our codebase to foster further research in the field of singing voice synthesis.

Voting advice applications (VAAs) help millions of voters understand which political parties or candidates best align with their views. This paper explores the potential risks these applications pose to the democratic process when targeted by adversarial entities. In particular, we expose 11 manipulation strategies and measure their impact using data from Switzerland’s primary VAA, Smartvote, collected during the last two national elections. We find that altering application parameters, such as the matching method, can shift a party’s recommendation frequency by up to 105%. Cherry-picking questionnaire items can increase party recommendation frequency by over 261%, while subtle changes to parties’ or candidates’ responses can lead to a 248% increase. To address these vulnerabilities, we propose adversarial robustness properties VAAs should satisfy, introduce empirical metrics for assessing the resilience of various matching methods, and suggest possible avenues for research toward mitigating the effect of manipulation. Our framework is key to ensuring secure and reliable AI-based VAAs poised to emerge in the near future.

The rise of reinforcement learning (RL) in critical real-world applications demands a fundamental rethinking of privacy in AI systems. Traditional privacy frameworks, designed to protect isolated data points, fall short for sequential decision-making systems where sensitive information emerges from temporal patterns, behavioral strategies, and collaborative dynamics. Modern RL paradigms, such as federated RL (FedRL) and RL with human feedback (RLHF) in large language models (LLMs), exacerbate these challenges by introducing complex, interactive, and context-dependent learning environments that traditional methods do not address. In this position paper, we argue for a new privacy paradigm built on four core principles: multi-scale protection, behavioral pattern protection, collaborative privacy preservation, and context-aware adaptation. These principles expose inherent tensions between privacy, utility, and interpretability that must be navigated as RL systems become more pervasive in high-stakes domains like healthcare, autonomous vehicles, and decision support systems powered by LLMs. To tackle these challenges, we call for the development of new theoretical frameworks, practical mechanisms, and rigorous evaluation methodologies that collectively enable effective privacy protection in sequential decision-making systems.

In this work, we investigate computational aspects of asynchronous cellular automata (ACAs), a modification of cellular automata in which cells update independently, following an asynchronous update schedule. We introduce flip automata networks (FANs), a simple modification of automata networks that remain robust under any asynchronous updating order. By using FANs as a middleman, we show that asynchronous automata can efficiently simulate their synchronous counterparts with a linear memory overhead, which improves upon the previously established quadratic bound. Additionally, we address the universality gap for (a)synchronous cellular automata-the boundary separating universal and non-universal automata, which is still not fully understood. We tighten this boundary by proving that all one-way asynchronous automata lack universal computational power. Conversely, we establish the existence of a universal asynchronous 6-state first-neighbor automaton in one dimension and a 3-state von Neumann automaton in two dimensions, which represent the smallest known universal constructions to date.

This work presents a first evaluation of two state-of-the-art Large Reasoning Models (LRMs), OpenAI’s o3-mini and DeepSeek R1, on analogical reasoning, focusing on well-established nonverbal human IQ tests based on Raven’s progressive matrices. We benchmark with the I-RAVEN dataset and its extension, I-RAVEN-X, which tests the ability to generalize to longer reasoning rules and ranges of the attribute values. To assess the influence of visual uncertainties on these symbolic analogical reasoning tests, we extend the I-RAVEN-X dataset, which otherwise assumes an oracle perception. We adopt a two-fold strategy to simulate this imperfect visual perception: 1) we introduce confounding attributes which, being sampled at random, do not contribute to the prediction of the correct answer of the puzzles, and 2) smooth the distributions of the input attributes’ values. We observe a sharp decline in OpenAI’s o3-mini task accuracy, dropping from 86.6% on the original I-RAVEN to just 17.0%—approaching random chance—on the more challenging I-RAVEN-X, which increases input length and range and emulates perceptual uncertainty. This drop occurred despite spending 3.4x more reasoning tokens. A similar trend is also observed for DeepSeek R1: from 80.6% to 23.2%. On the other hand, a neuro-symbolic probabilistic abductive model, ARLC, that achieves state-of-the-art performances on I-RAVEN, can robustly reason under all these out-of-distribution tests, maintaining strong accuracy with only a modest accuracy reduction from 98.6% to 88.0%. Our code is available at https://github.com/IBM/raven-large-language-models.

Compositional generalization—a key open challenge in modern machine learning—requires models to predict unknown combinations of known concepts. However, assessing compositional generalization remains a fundamental challenge due to the lack of standardized evaluation protocols and the limitations of current benchmarks, which often favor efficiency over rigor. At the same time, general-purpose vision architectures lack the necessary inductive biases, and existing approaches to endow them compromise scalability. As a remedy, this paper introduces: 1) a rigorous evaluation framework that unifies and extends previous approaches while reducing computational requirements from combinatorial to constant; 2) an extensive and modern evaluation on the status of compositional generalization in supervised vision backbones, training more than 5000 models; 3) Attribute Invariant Networks, a class of models establishing a new Pareto frontier in compositional generalization, achieving a 23.43% accuracy improvement over baselines while reducing parameter overhead from 600% to 16% compared to fully disentangled counterparts.

State-of-the-art medical multi-modal LLMs (med-MLLMs), such as LLaVA-Med and BioMedGPT, primarily depend on scaling model size and data volume, with training driven largely by autoregressive objectives. However, we reveal that this approach can lead to weak vision-language alignment, making these models overly dependent on costly instruction-following data. To address this, we introduce ExGra-Med, a novel multi-graph alignment framework that jointly aligns images, instruction responses, and extended captions in the latent space, advancing semantic grounding and cross-modal coherence. To scale to large LLMs (e.g., LLaMa-7B), we develop an efficient end-to-end training scheme using black-box gradient estimation, enabling fast and scalable optimization. Empirically, ExGra-Med matches LLaVA-Med’s performance using just 10\% of pre-training data, achieving a 20.13\% gain on VQA-RAD and approaching full-data performance. It also outperforms strong baselines like BioMedGPT and RadFM on visual chatbot and zero-shot classification tasks, demonstrating its promise for efficient, high-quality vision-language integration in medical AI.

Recent progress in speech processing has highlighted that high-quality performance across languages requires substantial training data for each individual language. While existing multilingual datasets cover many languages, they often contain insufficient data for each language, leading to models trained on these datasets to exhibit poor performance on most supported languages. Our work addresses this challenge by introducing a scalable pipeline for constructing speech datasets from parliamentary recordings. The proposed pipeline includes robust components for media retrieval and a two-stage alignment algorithm designed to handle non-verbatim transcripts and long-form audio. Applying this pipeline to recordings from 22 European parliaments, we extract over 61k hours of aligned speech segments, achieving substantial per-language coverage with 19 languages exceeding 1k hours and 22 languages exceeding 500 hours of high-quality speech data. We obtain an average 41.8\% reduction in word error rates over baselines when finetuning an existing ASR model on our dataset, demonstrating the usefulness of our approach.

Recent advances in 3D point cloud transformers have led to state-of-the-art results in tasks such as semantic segmentation and reconstruction. However, these models typically rely on dense token representations, incurring high computational and memory costs during training and inference. In this work, we present the finding that tokens are remarkably redundant, leading to substantial inefficiency. We introduce \textbf{GitMerge3D}, a \textbf{g}lobally \textbf{i}nformed graph \textbf{t}oken \textbf{merging} method that can reduce the token count by up to 90–95\% while maintaining competitive performance. This finding challenges the prevailing assumption that more tokens inherently yield better performance and highlights that many current models are over-tokenized and under-optimized for scalability. We validate our method across multiple 3D vision tasks and show consistent improvements in computational efficiency. This work is the first to assess redundancy in large-scale 3D transformer models, providing insights into the development of more efficient 3D foundation architectures. Our code and checkpoints are publicly available at \href{https://gitmerge3d.github.io/}{https://gitmerge3d.github.io}.

Generative models have made significant progress in synthesizing high-fidelity audio from short textual descriptions. However, editing existing audio using natural language has remained largely underexplored. Current approaches either require the complete description of the edited audio or are constrained to predefined edit instructions that lack flexibility. In this work, we introduce SAO-Instruct, a model based on Stable Audio Open capable of editing audio clips using any free-form natural language instruction. To train our model, we create a dataset of audio editing triplets (input audio, edit instruction, output audio) using Prompt-to-Prompt, DDPM inversion, and a manual editing pipeline. Although partially trained on synthetic data, our model generalizes well to real in-the-wild audio clips and unseen edit instructions. We demonstrate that SAO-Instruct achieves competitive performance on objective metrics and outperforms other audio editing approaches in a subjective listening study. To encourage future research, we release our code and model weights.

Modern neural networks exhibit state-of-the-art performance on many existing benchmarks, but their high computational requirements and energy usage cause researchers to explore more efficient solutions for real-world deployment. Differentiable logic gate networks (DLGNs) learns a large network of logic gates for efficient image classification. However, learning a network that can solve simple problems like CIFAR-10 or CIFAR-100 can take days to weeks to train. Even then, almost half of the neurons remains unused, causing a \emph{discretization gap}. This discretization gap hinders real-world deployment of DLGNs, as the performance drop between training and inference negatively impacts accuracy. We inject Gumbel noise with a straight-through estimator during training to significantly speed up training, improve neuron utilization, and decrease the discretization gap. We theoretically show that this results from implicit Hessian regularization, which improves the convergence properties of DLGNs. We train networks $4.5 \times$ faster in wall-clock time, reduce the discretization gap by 98\%, and reduce the number of unused gates by 100\%.

We study approximate agreement in an asynchronous network of n parties, up to t of which are byzantine. This an agreement task where the parties obtain approximately equal inputs in the convex hull of their inputs. In an asynchronous network, it can be solved with the optimal resilience t < n/3 by forcing the parties to reliably broadcast their messages and thus preventing inconsistent byzantine behavior. This costs Θ(n²) messages per reliable broadcast, or Θ(n³) messages per protocol iteration. In this work, we forgo reliable broadcast to achieve asynchronous approximate agreement against t < n/3 faults with quadratic communication. In a tree with the maximum degree Δ and the centroid decomposition height h, we achieve edge agreement (agreement on two adjacent vertices) in at most 6h + 1 rounds with 𝒪(n²) messages of size 𝒪(log Δ + log h) per round. We do this by designing a 6-round multivalued 2-graded consensus protocol and using it to construct a recursive edge agreement protocol. Then, we achieve edge agreement in the infinite path ℤ, again by using 2-graded consensus. Finally, we show that our edge agreement protocol enables approximate agreement in ℝ (with outputs that are at most some small parameter ε > 0 apart) in 6log₂M/(ε) + 𝒪(log log M/(ε)) rounds with 𝒪(n²) messages of size 𝒪(log log M/(ε)) per round, where M is the maximum non-byzantine input magnitude.

In stable matching, one must find a matching between two sets of agents, commonly men and women, or job applicants and job positions. Each agent has a preference ordering over who they want to be matched with. Moreover a matching is said to be stable if no pair of agents prefer each other over their current matching.

In this work, we study multivalued byzantine agreement (BA) in an asynchronous network of n parties where up to [EQUATION] parties are byzantine. We present a new reduction from multivalued BA to binary BA. It allows one to achieve BA on ℓ-bit inputs with one instance of binary BA, one instance of crusader agreement (CA) on ℓ-bit inputs and Θ(ℓn + n2) bits of additional communication.

Byzantine Agreement (BA) allows a set of n parties to agree on a value even when up to t of the parties involved are corrupted. While previous works have shown that, for ℓ-bit inputs, BA can be achieved with the optimal communication complexity O(ℓn) for sufficiently large ℓ, BA only ensures that honest parties agree on a meaningful output when they hold the same input, rendering the primitive inadequate for many real-world applications.

Byzantine Agreement (BA) considers a setting of $n$ parties, out of which up to $t$ can exhibit byzantine (malicious) behavior. Honest parties must decide on a common value (agreement), which must belong to a set determined by the honest inputs (validity). Depending on the use case, this set can grow or shrink, leading to various possible desiderata collectively known as validity conditions. Varying the validity property requirement can affect the regime under which BA is solvable. Our work investigates how the selected validity property impacts BA solvability in the network-agnostic model, where the network can either be synchronous with up to $t_s$ byzantine parties or asynchronous with up to $t_a \leq t_s$ byzantine parties. We give necessary and sufficient conditions for a validity property to render BA solvable, both for the case with cryptographic setup and for the one without. This traces the precise boundary of solvability in the network-agnostic model for every validity property. Our proof of sufficiency provides a universal protocol, that achieves BA for a given validity property whenever the provided conditions are satisfied. We note that, for any non-trivial validity property, the condition $2 \cdot t_s + t_a < n$ is necessary for BA to be solvable, even with cryptographic setup. Specializing this claim to $t_a = 0$ gives that $t < n / 2$ is required whenever one expects a purely synchronous protocol to also work in an asynchronous network when there are no corruptions. This is especially surprising given that, for some validity properties, $t < n$ is a sufficient condition without the last stipulation.

We study the problem of Strong Byzantine Agreement and establish tight upper and lower bounds on communication complexity, parameterized by the actual number of Byzantine faults. Specifically, for a system of n parties tolerating up to t Byzantine faults, out of which only f ≤ t are actually faulty, we obtain the following results: In the partially synchronous setting, we present the first Byzantine Agreement protocol that achieves adaptive communication complexity of 𝒪(n + t ⋅ f) words, which is asymptotically optimal. Our protocol has an optimal resilience of t < n/3. In the asynchronous setting, we prove a lower bound of Ω(n + t²) on the expected number of messages, and design an almost matching protocol with an optimal resilience that solves agreement with 𝒪((n + t²)⋅ log n) words. Our main technical contribution in the asynchronous setting is the utilization of a bipartite expander graph that allows for low-cost information dissemination.

We consider an asynchronous network of n message-sending parties, up to t of which are byzantine. We study approximate agreement, where the parties obtain approximately equal outputs in the convex hull of their inputs. In their seminal work, Abraham, Amit and Dolev [OPODIS '04] solve this problem in ℝ with the optimal resilience t < n/3 with a protocol where each party reliably broadcasts a value in every iteration. This takes Θ(n²) messages per reliable broadcast, or Θ(n³) messages per iteration. In this work, we forgo reliable broadcast to achieve asynchronous approximate agreement against t < n/3 faults with quadratic communication. In a tree with the maximum degree Δ and the centroid decomposition height h, we achieve edge agreement in at most 6h + 1 rounds with 𝒪(n²) messages of size 𝒪(log Δ + log h) per round. We do this by designing a 6-round multivalued 2-graded consensus protocol and using it to recursively reduce the task to edge agreement in a subtree with a smaller centroid decomposition height. Then, we achieve edge agreement in the infinite path ℤ, again with the help of 2-graded consensus. Finally, we show that our edge agreement protocol enables ε-agreement in ℝ in 6log₂M/(ε) + 𝒪(log log M/(ε)) rounds with 𝒪(n² log M/(ε)) messages and 𝒪(n²log M/(ε)log log M/(ε)) bits of communication, where M is the maximum non-byzantine input magnitude.

This paper analyses a set of simple adaptations that transform standard message-passing Graph Neural Networks (GNN) into provably powerful directed multigraph neural networks. The adaptations include multigraph port numbering, ego IDs, and reverse message passing. We prove that the combination of these theoretically enables the detection of any directed subgraph pattern. To validate the effectiveness of our proposed adaptations in practice, we conduct experiments on synthetic subgraph detection tasks, which demonstrate outstanding performance with almost perfect results. Moreover, we apply our proposed adaptations to two financial crime analysis tasks. We observe dramatic improvements in detecting money laundering transactions, improving the minority-class F1 score of a standard message-passing GNN by up to 30%, and closely matching or outperforming tree-based and GNN baselines. Similarly impressive results are observed on a real-world phishing detection dataset, boosting three standard GNNs’ F1 scores by around 15% and outperforming all baselines. An extended version with appendices can be found on arXiv: https://arxiv.org/abs/2306.11586.

Decentralized Autonomous Organizations (DAOs) are blockchain-based organizations that facilitate decentralized governance. Today, DAOs not only hold billions of dollars in their treasury but also govern many of the most popular Decentralized Finance (DeFi) protocols. This paper systematically analyses security threats to DAOs, focusing on the types of attacks they face. We study attacks on DAOs that took place in the past, attacks that have been theorized to be possible, and potential attacks that were uncovered and prevented in audits. For each of these (potential) attacks, we describe and categorize the attack vectors utilized into four categories. This reveals that while many attacks on DAOs take advantage of the less tangible and more complex human nature involved in governance, audits tend to focus on code and protocol vulnerabilities. Thus, additionally, the paper examines empirical data on DAO vulnerabilities, outlines risk factors contributing to these attacks, and suggests mitigation strategies to safeguard against such vulnerabilities.

Efficient (non-interactive) aggregate signature schemes in the post-quantum setting are still missing. We define the concept of Byzantine fault-tolerant aggregate signature schemes. This notion fills a middle-ground between interactive and non-interactive signature aggregation. Signers are required to interact but some may fail or misbehave at any point in the signing protocol.

Our work examines the efficacy of employing advanced machine learning methods to solve captchas from Google's reCAPTCHAv2 system. We evaluate the effectiveness of automated systems in solving captchas by utilizing advanced YOLO models for image segmentation and classification. Our main result is that we can solve 100% of the captchas, while previous work only solved 68–71 %. Furthermore, our findings suggest that there is no significant difference in the number of challenges humans and bots must solve to pass the captchas in reCAPTCHAv2. This implies that current AI technologies can exploit advanced image-based captchas. We also look under the hood of reCAPTCHAv2, and find evidence that reCAPTCHAv2 is heavily based on cookie and browser history data when evaluating whether a user is human or not. The code is provided alongside this paper.11https://github.com/aplesner/Breaking-reCAPTCHAv2

Reinforcement learning (RL) has gained popularity in the realm of recommender systems due to its ability to optimize long-term rewards and guide users in discovering relevant content. However, the successful implementation of RL in recommender systems is challenging because of several factors, including the limited availability of online data for training on-policy methods. This scarcity requires expensive human interaction for online model training. Furthermore, the development of effective evaluation frameworks that accurately reflect the quality of models remains a fundamental challenge in recommender systems. To address these challenges, we propose a comprehensive framework for synthetic environments that simulate human behavior by harnessing the capabilities of large language models (LLMs). We complement our framework with in-depth ablation studies and demonstrate its effectiveness with experiments on movie and book recommendations. Using LLMs as synthetic users, this work introduces a modular and novel framework to train RL-based recommender systems. The software, including the RL environment, is publicly available on https://github.com/SUBER-Team/SUBER.

Many graph algorithms can be viewed as sets of rules that are iteratively applied, with the number of iterations dependent on the size and complexity of the input graph. Existing machine learning architectures often struggle to represent these algorithmic decisions as discrete state transitions. Therefore, we propose a novel framework: GraphFSA (Graph Finite State Automaton). GraphFSA is designed to learn a finite state automaton that runs on each node of a given graph. We test GraphFSA on cellular automata problems, showcasing its abilities in a straightforward algorithmic setting. For a comprehensive empirical evaluation of our framework, we create a diverse range of synthetic problems. As our main application, we then focus on learning more elaborate graph algorithms. Our findings suggest that GraphFSA exhibits strong generalization and extrapolation abilities, presenting an alternative approach to represent these algorithms.

Serverless ML inference is an emerging cloud computing paradigm for low-cost, easy-to-manage inference services. In serverless ML inference, each call is executed in a container; however, the cold start of containers results in long inference delays. Unfortunately, most existing works do not work well because they still need to load models into containers from scratch, which is the bottleneck based on our observations. Therefore, this paper proposes a low-latency serverless ML inference system called Optimus via a new container management scheme. Our key insight is that the loading of a new model can be significantly accelerated when using an existing model with a similar structure in a warm but idle container. We thus develop a novel idea of inter-function model transformation for serverless ML inference, which delves into models within containers at a finer granularity of operations, designs a set of in-container meta-operators for both CNN and transformer model transformation, and develops an efficient scheduling algorithm with linear complexity for a low-cost transformation strategy. Our evaluations on thousands of models show that Optimus reduces inference latency by 24.00% ~ 47.56% in both simulated and real-world workloads compared to state-of-the-art work.

In the realm of generative models for graphs, extensive research has been conducted. However, most existing methods struggle with large graphs due to the complexity of representing the entire joint distribution across all node pairs and capturing both global and local graph structures simultaneously. To overcome these issues, we introduce a method that generates a graph by progressively expanding a single node to a target graph. In each step, nodes and edges are added in a localized manner through denoising diffusion, building first the global structure, and then refining the local details. The local generation avoids modeling the entire joint distribution over all node pairs, achieving substantial computational savings with subquadratic runtime relative to node count while maintaining high expressivity through multiscale generation. Our experiments show that our model achieves state-of-the-art performance on well-established benchmark datasets while successfully scaling to graphs with at least 5000 nodes. Our method is also the first to successfully extrapolate to graphs outside of the training distribution, showcasing a much better generalization capability over existing methods.

Graph Visualization, also known as Graph Drawing, aims to find geometric embeddings of graphs that optimize certain criteria. Stress is a widely used metric; stress is minimized when every pair of nodes is positioned at their shortest path distance. However, stress optimization presents computational challenges due to its inherent complexity and is usually solved using heuristics in practice. We introduce a scalable Graph Neural Network (GNN) based Graph Drawing framework with sub-quadratic runtime that can learn to optimize stress. Inspired by classical stress optimization techniques and force-directed layout algorithms, we create a coarsening hierarchy for the input graph. Beginning at the coarsest level, we iteratively refine and un-coarsen the layout, until we generate an embedding for the original graph. To enhance information propagation within the network, we propose a novel positional rewiring technique based on intermediate node positions. Our empirical evaluation demonstrates that the framework achieves state-of-the-art performance while remaining scalable.

We propose a new self-explainable Graph Neural Network (GNN) model: GraphChef. GraphChef integrates decision trees into the GNN message passing framework. Given a dataset, GraphChef returns a set of rules (a recipe) that explains each class in the dataset unlike existing GNNs and explanation methods that reason on individual graphs. Thanks to the decision trees, GraphChef recipes are human understandable. We also present a new pruning method to produce small and easy to digest trees. Experiments demonstrate that GraphChef reaches comparable accuracy to not self-explainable GNNs and produced decision trees are indeed small. We further validate the correctness of the discovered recipes on datasets where explanation ground truth is available: Reddit-Binary, MUTAG, BA-2Motifs, BA-Shapes, Tree-Cycle, and Tree-Grid.

This paper presents Banyan, the first rotating leader state machine replication (SMR) protocol that allows transactions to be confirmed in just a single round-trip time in the Byzantine fault tolerance (BFT) setting. Based on minimal alterations to the Internet Computer Consensus (ICC) protocol and with negligible communication overhead, we introduce a novel dual mode mechanism that enables optimal block finalization latency in the fast path. Crucially, the modes of operation are integrated, such that even if the fast path is not effective, no penalties are incurred. Moreover, our algorithm maintains the core attributes of the ICC protocol it is based on, including optimistic responsiveness and rotating leaders without the necessity for a view-change protocol.

While autonomous agents often surpass humans in their ability to handle vast and complex data, their potential misalignment (i.e., lack of transparency regarding their true objective) has thus far hindered their use in critical applications such as social decision processes. More importantly, existing alignment methods provide no formal guarantees on the safety of such models. Drawing from utility and social choice theory, we provide a novel quantitative definition of alignment in the context of social decision-making. Building on this definition, we introduce probably approximately aligned (i.e., near-optimal) policies, and we derive a sufficient condition for their existence. Lastly, recognizing the practical difficulty of satisfying this condition, we introduce the relaxed concept of safe (i.e., nondestructive) policies, and we propose a simple yet robust method to safeguard the black-box policy of any autonomous agent, ensuring all its actions are verifiably safe for the society.

Algorithmic reasoning is a fundamental cognitive ability that plays a pivotal role in problem-solving and decision-making processes. Reinforcement Learning (RL) has demonstrated remarkable proficiency in tasks such as motor control, handling perceptual input, and managing stochastic environments. These advancements have been enabled in part by the availability of benchmarks. In this work we introduce PUZZLES, a benchmark based on Simon Tatham's Portable Puzzle Collection, aimed at fostering progress in algorithmic and logical reasoning in RL. PUZZLES contains 40 diverse logic puzzles of adjustable sizes and varying levels of complexity, providing detailed information on the strengths and generalization capabilities of RL agents. Furthermore, we evaluate various RL algorithms on PUZZLES, providing baseline comparisons and demonstrating the potential for future research. All the software, including the environment, is available at this https url.

Byzantine Agreement (BA) allows a set of n parties to agree on a value even when up to t of the parties involved are corrupted. While previous works have shown that, for ℓ-bit inputs, BA can be achieved with the optimal communication complexity Õ(ℓn) for sufficiently large ℓ, BA only ensures that honest parties agree on a meaningful output when they hold the same input, rendering the primitive inadequate for many real-world applications.

Image datasets serve as the foundation for machine learning models in computer vision, significantly influencing model capabilities, performance, and biases alongside architectural considerations. Therefore, understanding the composition and distribution of these datasets has become increasingly crucial. To address the need for intuitive exploration of these datasets, we propose AEye, an extensible and scalable visualization tool tailored to image datasets. AEye utilizes a contrastively trained model to embed images into semantically meaningful high-dimensional representations, facilitating data clustering and organization. To visualize the high-dimensional representations, we project them onto a two-dimensional plane and arrange images in layers so users can seamlessly navigate and explore them interactively. AEye facilitates semantic search functionalities for both text and image queries, enabling users to search for content. We open-source the codebase for AEye, and provide a simple configuration to add datasets.

We empirically study the state of three prominent DAO governance systems on the Ethereum blockchain: Compound, Uniswap and ENS. In particular, we examine how the voting power is distributed in these systems. Using a comprehensive dataset of all governance token holders, delegates, proposals and votes, we analyze who holds the voting power and how this power is being used to influence governance decisions. While we reveal that the majority of voting power is concentrated in the hands of a small number of addresses, we rarely observe these powerful entities overturning a vote by choosing a different outcome than that of the overall community and less influential voters.

Combinatorial auctions (CAs) offer the flexibility for bidders to articulate complex preferences when competing for multiple assets. However, the behavior of bidders under different payment rules is often unclear. Our research explores the relationship between core constraints and several core-selecting payment rules. Specifically, we examine the natural and desirable property of payment rules of being non-decreasing, which ensures that bidding higher does not lead to lower payments. Earlier studies revealed that the VCG-nearest payment method – a commonly employed payment rule – fails to adhere to this principle even for single-minded CAs. We establish that when a single effective core constraint exists, the payment maintains the non-decreasing property in single-minded CAs. To identify auctions where such a constraint is present, we introduce a novel framework using conflict graphs to represent single-minded CAs and establish sufficient conditions for the existence of single effective core constraints. We proceed with an analysis of the implications on bidder behavior, demonstrating that there is no overbidding in any Nash equilibrium when considering non-decreasing core-selecting payment rules. Our study concludes by establishing the non-decreasing nature of two additional payment rules, namely the proxy and proportional payment rules, for single-minded CAs.

In this work, we investigate the robustness of BERT using four word substitution-based attacks. We combine a human evaluation of individual word substitutions and probabilistic analysis to show that most of the adversarial examples from the four studied attacks do not preserve enough semantics from the original examples, and can thus be easily recognized by human annotators. To further confirm that, we introduce an efficient adversarial defense consisting of a data augmentation step and a post-processing step. We show that many successful attacks can be defended using our defense method by including data similar to adversarial examples during training.

We study the problem of learning comparisons between numbers with neural networks. Despite comparisons being a seemingly simple problem, we find that both general-purpose models such as multilayer perceptrons (MLPs) as well as arithmetic architectures such as the Neural Arithmetic Logic Unit (NALU) struggle with learning comparisons. Neither architecture can extrapolate to much larger numbers than those seen in the training set. We propose a novel differentiable architecture, the Neural Status Register (NSR) to solve this problem. We experimentally validate the NSR in various settings. We can combine the NSR with other neural models to solve interesting problems such as piecewise-defined arithmetic, comparison of digit images, recurrent problems, or finding shortest paths in graphs. The NSR outperforms all baseline architectures, especially when it comes to extrapolating to larger numbers.

In the evolving landscape of digital art, NonFungible Tokens (NFTs) have emerged as a groundbreaking platform, bridging the realms of art and technology. NFTs serve as the foundational framework that has revolutionized the market for digital art, enabling artists to showcase and monetize their creations in unprecedented ways. NFTs combine metadata stored on the blockchain with off-chain data, such as images, to create a novel form of digital ownership. It is not fully understood how these factors come together to determine NFT prices. In this study, we analyze both on-chain and off-chain data of NFT collections trading on OpenSea to understand what influences NFT pricing. Our results show that while text and image data of the NFTs can be used to explain price variations within collections, the extracted features do not generalize to new, unseen collections. Furthermore, we find that an NFT collection's trading volume often relates to its online presence, like social media followers and website traffic.

Rigid origami has shown potential in large diversity of practical applications. However, current rigid origami crease pattern design mostly relies on known tessellations. This strongly limits the diversity and novelty of patterns that can be created. In this work, we build upon the recently developed principle of three units method to formulate rigid origami design as a discrete optimization problem, the rigid origami game. Our implementation allows for a simple definition of diverse objectives and thereby expands the potential of rigid origami further to optimized, application-specific crease patterns. We showcase the flexibility of our formulation through use of a diverse set of search methods in several illustrative case studies. We are not only able to construct various patterns that approximate given target shapes, but to also specify abstract, function-based rewards which result in novel, foldable and functional designs for everyday objects.

With Ethereum's transition from Proof-of-Work to Proof-of-Stake in September 2022 came another paradigm shift, the Proposer-Builder Separation (PBS) scheme. PBS was introduced to decouple the roles of selecting and ordering transactions in a block (i.e., the builder), from those validating its contents and proposing the block to the network as the new head of the blockchain (i.e., the proposer). In this landscape, proposers are the validators in the Proof-of-Stake consensus protocol, while now relying on specialized block builders for creating blocks with the highest value for the proposer. Additionally, relays act as mediators between builders and proposers. We study PBS adoption and show that the current landscape exhibits significant centralization amongst the builders and relays. Further, we explore whether PBS effectively achieves its intended objectives of enabling hobbyist validators to maximize block profitability and preventing censorship. Our findings reveal that although PBS grants validators the opportunity to access optimized and competitive blocks, it tends to stimulate censorship rather than reduce it. Additionally, we demonstrate that relays do not consistently uphold their commitments and may prove unreliable. Specifically, proposers do not always receive the complete promised value, and the censorship or filtering capabilities pledged by relays exhibit significant gaps.

Music datasets play a crucial role in advancing research in machine learning for music. However, existing music datasets suffer from limited size, accessibility, and lack of audio resources. To address these shortcomings, we present DISCO-10M, a novel and extensive music dataset that surpasses the largest previously available music dataset by an order of magnitude. To ensure high-quality data, we implement a multi-stage filtering process. This process incorporates similarities based on textual descriptions and audio embeddings. Moreover, we provide precomputed CLAP embeddings alongside DISCO-10M, facilitating direct application on various downstream tasks. These embeddings enable efficient exploration of machine learning applications on the provided data. With DISCO-10M, we aim to democratize and facilitate new research to help advance the development of novel machine learning models for music: https://huggingface.co/DISCOX

Traditional blockchain design gives miners or validators full control over transaction ordering, i.e., they can freely choose which transactions to include or exclude, as well as in which order. While not an issue initially, the emergence of decentralized finance has introduced new transaction order dependencies allowing parties in control of the ordering to make a profit by front-running others' transactions. In this work, we present the Decentralized Clock Network, a new approach for achieving fair transaction ordering. Users submit their transactions to the network's clocks, which run an agreement protocol that provides each transaction with a timestamp of receipt which is then used to define the transactions' order. By separating agreement from ordering, our protocol is efficient and has a simpler design compared to other available solutions. Moreover, our protocol brings to the blockchain world the paradigm of asynchronous fallback, where the algorithm operates with stronger fairness guarantees during periods of synchronous use, switching to an asynchronous mode only during times of increased network delay.

In the realm of computer science, it may seem that distributed computing and machine learning exist on opposite ends of the spectrum. However, there are many connections between the two domains, both in theory and practice.

Within just four years, the blockchain-based Decentralized Finance (DeFi) ecosystem has accumulated a peak total value locked (TVL) of more than 253 billion USD. This surge in DeFi’s popularity has, unfortunately, been accompanied by many impactful incidents. According to our data, users, liquidity providers, speculators, and protocol operators suffered a total loss of at least 3.24 billion USD from Apr 30, 2018 to Apr 30, 2022. Given the blockchain’s transparency and increasing incident frequency, two questions arise: How can we systematically measure, evaluate, and compare DeFi incidents? How can we learn from past attacks to strengthen DeFi security?In this paper, we introduce a common reference frame to systematically evaluate and compare DeFi incidents, including both attacks and accidents. We investigate 77 academic papers, 30 audit reports, and 181 real-world incidents. Our data reveals several gaps between academia and the practitioners’ community. For example, few academic papers address "price oracle attacks" and "permissonless interactions", while our data suggests that they are the two most frequent incident types (15% and 10.5% correspondingly). We also investigate potential defenses, and find that: (i) 103 (56%) of the attacks are not executed atomically, granting a rescue time frame for defenders; (ii) bytecode similarity analysis can at least detect 31 vulnerable/23 adversarial contracts; and (iii) 33 (15.3%) of the adversaries leak potentially identifiable information by interacting with centralized exchanges.

Multidimensional Approximate Agreement considers a setting of n parties, where each party holds a vector in ℝD as input. The honest parties are required to obtain very close outputs in ℝD that lie inside the convex hull of their inputs.

This paper studies the question whether automated market maker protocols such as Uniswap can sustainably retain a portion of their trading fees for the protocol. We approach the problem by modelling how to optimally choose a pool's take rate, i.e the fraction of fee revenue that remains with the protocol, in order to maximize the protocol's revenue. The model suggest that if AMMs have a portion of loyal trade volume, they can sustainably set a non-zero take rate, even without losing liquidity to competitors with a zero take rate. Furthermore, we determine the optimal take rate depending on a number of model parameters including how much loyal trade volume pools have and how high the competitors' take rates are.

Trade execution on Decentralized Exchanges (DEXes) is automatic and does not require individual buy and sell orders to be matched. Instead, liquidity aggregated in pools from individual liquidity providers enables trading between cryptocurrencies. The largest DEX measured by trading volume, Uniswap V3, promises a DEX design optimized for capital efficiency. However, Uniswap V3 requires far more decisions from liquidity providers than previous DEX designs.

User transactions on Ethereum's peer-to-peer network are at risk of being attacked. The smart contracts building decentralized finance (DeFi) have introduced a new transaction ordering dependency to the Ethereum blockchain. As a result, attackers can profit from front- and back-running transactions. Multiple approaches to mitigate transaction reordering manipulations have surfaced recently. However, the success of individual approaches in mitigating such attacks and their impact on the entire blockchain remains largely unstudied.

Predatory trading bots lurking in Ethereum's mempool present invisible taxation of traders on automated market makers (AMMs). AMM traders specify a slippage tolerance to indicate the maximum price movement they are willing to accept. This way, traders avoid automatic transaction failure in case of small price movements before their trade request executes. However, while a too-small slippage tolerance may lead to trade failures, a too-large slippage tolerance allows predatory trading bots to profit from sandwich attacks. These bots can extract the difference between the slippage tolerance and the actual price movement as profit.

Decentralized finance (DeFi) enables crypto-asset holders to conduct complex financial transactions, while maintaining control over their assets in the blockchain ecosystem. However, the transparency of blockchain networks and the open mechanism of DeFi applications also cause new security issues. In this paper, we focus on sandwich attacks, where attackers take advantage of the transaction confirmation delay and cause financial losses for victims. We evaluate the impact and investigate users’ perceptions of sandwich attacks through a mix-method study. We find that due to users’ lack of technical background and insufficient notifications from the markets, many users were not aware of the existence and the impact of sandwich attacks. They also had a limited understanding of how to resolve the security issue. Interestingly, users showed high tolerance for the impact of sandwich attacks on individuals and the ecosystem, despite potential financial losses. We discuss general implications for users, DeFi applications, and the community.

In this work, we introduce a novel pipeline that enables the generation of multiple-choice questions and exercises from as little as a topic keyword. Hence, providing users the possibility to start with a study objective in mind and then automatically generate personalized learning material. The main contributions of this project are a scraper that can extract relevant information from websites, a novel distractor generation method that can make use of context and a technique to automatically combine text and questions into interactive exercises. Our novel distractor generation method was tested in a human survey which showed that the distractor generation quality is comparable to hand crafted distractors. The pipeline is built into a web application that lets users refine the results for each step, openly accessible at https://adaptive-teaching.com.

Recent work has demonstrated that pre-trained language models (PLMs) are zero-shot learners.However, most existing zero-shot methods involve heavy human engineering or complicated self-training pipelines, hindering their application to new situations.In this work, we show that zero-shot text classification can be improved simply by clustering texts in the embedding spaces of PLMs.Specifically, we fit the unlabeled texts with a Bayesian Gaussian Mixture Model after initializing cluster positions and shapes using class names.Despite its simplicity, this approach achieves superior or comparable performance on both topic and sentiment classification datasets and outperforms prior works significantly on unbalanced datasets.We further explore the applicability of our clustering approach by evaluating it on 14 datasets with more diverse topics, text lengths, and numbers of classes.Our approach achieves an average of 20% absolute improvement over prompt-based zero-shot learning.Finally, we compare different PLM embedding spaces and find that texts are well-clustered by topics even if the PLM is not explicitly pre-trained to generate meaningful sentence embeddings.This work indicates that PLM embeddings can categorize texts without task-specific fine-tuning, thus providing a new way to analyze and utilize their knowledge and zero-shot learning ability 1 .

Brain decoding, understood as the process of mapping brain activities to the stimuli that generated them, has been an active research area in the last years. In the case of language stimuli, recent studies have shown that it is possible to decode fMRI scans into an embedding of the word a subject is reading. However, such word embeddings are designed for natural language processing tasks rather than for brain decoding. Therefore, they limit the model’s ability to recover the precise stimulus. In this work, we propose to directly classify an fMRI scan, mapping it to the corresponding word within a fixed vocabulary. Unlike existing work, we evaluate on scans from previously unseen subjects. We argue that this is a more realistic setup and we present a model that can decode fMRI data from unseen subjects with 2.62% Top-1 and 9.76% Top5 accuracy in this challenging task. Moreover our model can be fine-tuned on data from the test subject to achieve 4.22% Top-1 and 12.87% Top-5 accuracy, significantly outperforming all the considered competitive baselines.

Blockchain "on-chain" oracles are critical to the functioning of many Decentralized Finance (DeFi) protocols. We analyze these oracles for manipulation resistance. Specifically, we analyze the cost of manipulating on-chain time-weighted average price (TWAP) oracles that use the arithmetic mean. It has been assumed that manipulating a TWAP oracle with the well-known multi-block attack is expensive and scales linearly with the length of the TWAP. We question this assumption with two novel results. First, we describe a single-block attack that works under the same setting as the multi-block attack but costs less to execute. Second, we describe a multi-block MEV (MMEV) style attack where the attacker colludes with a miner/proposer who can mine/propose two blocks in a row. This MMEV style attack makes oracle manipulation orders of magnitude cheaper than previously known attacks. In the proof-of-work setting, MMEV can be done by selfish mining even with very low shares of hashpower.

Atomic Swaps enable exchanging crypto-assets with-out trusting a third party. To enable these swaps, both parties lock funds and let their counterparty withdraw them in exchange for a secret. This leads to the so-called griefing attack, or the emergence of an American Call option, where one party stops participating in the swap, thereby making their counterparty wait for a timelock to expire before they can withdraw their funds. The standard way to mitigate this attack is to make the attacker pay a premium for the emerging American Call option. In these premium-paying approaches, the premium itself ends up being locked for possibly an even longer duration than the swap amount itself. We propose a new Atomic Swap construction, where neither party exposes itself to a griefing attack by their counterparty. Notably, unlike previous constructions, ours can be implemented in Bitcoin as is. Our construction also takes fewer on-chain transactions and has a lower worst-case timelock.

We approach the graph generation problem from a spectral perspective by first generating the dominant parts of the graph Laplacian spectrum and then building a graph matching these eigenvalues and eigenvectors. Spectral conditioning allows for direct modeling of the global and local graph structure and helps to overcome the expressivity and mode collapse issues of one-shot graph generators. Our novel GAN, called SPECTRE, enables the one-shot generation of much larger graphs than previously possible with one-shot models. SPECTRE outperforms state-of-the-art deep autoregressive generators in terms of modeling fidelity, while also avoiding expensive sequential generation and dependence on node ordering. A case in point, in sizable synthetic and real-world graphs SPECTRE achieves a 4-to-170 fold improvement over the best competitor that does not overfit and is 23-to-30 times faster than autoregressive generators.

We study and compare different Graph Neural Network extensions that increase the expressive power of GNNs beyond the Weisfeiler-Leman test. We focus on (i) GNNs based on higher order WL methods, (ii) GNNs that preprocess small substructures in the graph, (iii) GNNs that preprocess the graph up to a small radius, and (iv) GNNs that slightly perturb the graph to compute an embedding. We begin by presenting a simple improvement for this last extension that strictly increases the expressive power of this GNN variant. Then, as our main result, we compare the expressiveness of these extensions to each other through a series of example constructions that can be distinguished by one of the extensions, but not by another one. We also show negative examples that are particularly challenging for each of the extensions, and we prove several claims about the ability of these extensions to count cliques and cycles in the graph.

The collection of eye gaze information provides a window into many critical aspects of human cognition, health and behaviour. Additionally, many neuroscientific studies complement the behavioural information gained from eye tracking with the high temporal resolution and neurophysiological markers provided by electroencephalography (EEG). One of the essential eye-tracking software processing steps is the segmentation of the continuous data stream into events relevant to eye-tracking applications, such as saccades, fixations, and blinks. Here, we introduce DETRtime, a novel framework for time-series segmentation that creates ocular event detectors that do not require additionally recorded eye-tracking modality and rely solely on EEG data. Our end-to-end deep-learning-based framework brings recent advances in Computer Vision to the forefront of the times series segmentation of EEG data. DETRtime achieves state-of-the-art performance in ocular event detection across diverse eye-tracking experiment paradigms. In addition to that, we provide evidence that our model generalizes well in the task of EEG sleep stage segmentation.

Grammatical inference is a classical problem in computational learning theory and a topic of wider influence in natural language processing. We treat grammars as a model of computation and propose a novel neural approach to induction of regular grammars from positive and negative examples. Our model is fully explainable, its intermediate results are directly interpretable as partial parses, and it can be used to learn arbitrary regular grammars when provided with sufficient data. We find that our method consistently attains high recall and precision scores across a range of tests of varying complexity.

We introduce two-crossing elections as a generalization of single-crossing elections, showing a number of new results. First, we show that two-crossing elections can be recognized in polynomial time, by reduction to the well-studied consecutive ones problem. Single-crossing elections exhibit a transitive majority relation, from which many important results follow. On the other hand, we show that the classical Debord-McGarvey theorem can still be proven two-crossing, implying that any weighted majority tournament is inducible by a two-crossing election. This shows that many voting rules are NP-hard under two-crossing elections, including Kemeny and Slater. This is in contrast to the single-crossing case and outlines an important complexity boundary between single- and two-crossing. Subsequently, we show that for two-crossing elections the Young scores of all candidates can be computed in polynomial time, by formulating a totally unimodular linear program. Finally, we consider the Chamberlin-Courant rule with arbitrary disutilities and show that a winning committee can be computed in polynomial time, using an approach based on dynamic programming.

Integer sequences are of central importance to the modeling of concepts admitting complete finitary descriptions. We introduce a novel view on the learning of such concepts and lay down a set of benchmarking tasks aimed at conceptual understanding by machine learning models. These tasks indirectly assess model ability to abstract, and challenge them to reason both interpolatively and extrapolatively from the knowledge gained by observing representative examples. To further aid research in knowledge representation and reasoning, we present FACT, the Finitary Abstraction Comprehension Toolkit. The toolkit surrounds a large dataset of integer sequences comprising both organic and synthetic entries, a library for data pre-processing and generation, a set of model performance evaluation tools, and a collection of baseline model implementations, enabling the making of the future advancements with ease.

Approximate Agreement (AA) allows a set of n parties that start with real-valued inputs to obtain values that are at most within a parameter ε > 0 from each other and within the range of their inputs. Existing AA protocols, both for the synchronous network model (where any message is delivered within a known delay Δ time) and the asynchronous network model, are secure when up to t < n/3 of the parties are corrupted and require no initial setup (such as a public-key infrastructure (PKI) for signatures). We consider AA protocols where a PKI is available, and show the first AA protocol that achieves simultaneously security against ts corruptions when the network is synchronous and ta corruptions when the network is asynchronous, for any 0 ≤ ta < n/3 ≤ ts < n/2 such that ta + 2 · ts < n. We further show that our protocol is optimal by proving that achieving AA for ta +2·ts ≥ n is impossible (even with setup). Remarkably, this is also the first AA protocol that tolerates more than n/3 corruptions in the synchronous network model.

Abstract Live Streaming Commerce (LSC) is proliferating in China and gaining traction worldwide. LSC is an e-commerce service where sellers communicate with consumers through live streaming while consumers can place orders within the same system. Despite the significant involvement of consumers in LSC, it has not been systematically analyzed how consumers make shopping decisions when engaging with LSC. In this paper, we conduct a mixed-methods study, consisting of surveys ( N 1 = 240) and follow-up interviews ( N 2 = 16) with LSC consumers. We focus on two features of LSC, i.e., the communication between merchants and consumers through live streaming and the participation of streamers, and aim to understand how these changes influence consumers’ decision-making process in LSC. We find that LSC enables merchants to exchange information with consumers based on their needs and provide additional customer services. Because of the appropriate information about the products they acquire and the enjoyable shopping atmosphere, consumers are willing to purchase products in LSC. As the intermediaries between merchants and consumers, streamers utilize their independent identity from merchants to enhance consumers’ awareness of shopping and persuade their online shopping decisions. Moreover, we consider the opportunities and challenges of current LSC services and provide implications for LSC services and the research community regarding the development of LSC.

Gay dating applications, such as Grindr and SCRUFF, are considered the primary platforms for gay men to conduct online dating activities. However, on Zhihu, a Chinese question-and-answer website, tens of thousands of homosexual users have been searching for romantic partners, which suggests that Zhihu may have unique affordances in online dating activities for Chinese gay men. To better understand how Chinese gay men perceive the affordances of a non-dating platform for online dating, we conduct a mixed-methods study, including observations, interviews, and quantitative and qualitative analysis of users' self-presentations. We find that gay men users publish personal ads by answering "fishing questions" on Zhihu. Through our analysis, we examine how users perceive the affordances of Zhihu to satisfy their social and psychological gratifications at the self, community, and audience levels. Although gay users face the risk of disclosing homosexual identity on mainstream social media, they perceive such risk as acceptable for better online dating experience. We discuss how users respond to severe social stigma in China, and the gap between user needs and the design of gay dating applications. We elaborate on the implications of our findings to discuss the potential benefits for LGBTQ users if LGBTQ service providers collaborate with social media.

Yiran Xing, Zai Shi, Zhao Meng, Gerhard Lakemeyer, Yunpu Ma, Roger Wattenhofer. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

We take a deep look into the behaviour of selfattention heads in the transformer architecture.In light of recent work discouraging the use of attention distributions for explaining a model's behaviour, we show that attention distributions can nevertheless provide insights into the local behaviour of attention heads.This way, we propose a distinction between local patterns revealed by attention and global patterns that refer back to the input, and analyze BERT from both angles.We use gradient attribution to analyze how the output of an attention head depends on the input tokens, effectively extending the local attention-based analysis to account for the mixing of information throughout the transformer layers.We find that there is a significant mismatch between attention and attribution distributions, caused by the mixing of context inside the model.We quantify this discrepancy and observe that interestingly, there are some patterns that persist across all layers despite the mixing.

Speech enhancement is one of the biggest challenges in hearing prosthetics. In face-to-face communication devices have to estimate the signal of interest, but playback of speech signals from an electronic device opens up new opportunities. Audio signals can be enriched with hidden data, which can subsequently be decoded by the receiver. We investigate a hybrid strategy made of signal processing and RNN (Recurrent Neural Networks) to calculate and compress cepstral coefficients: these are descriptors of the speech signal, which can be embedded in the signal itself and used at the receiver’s end to perform an Informed Speech Enhancement. Objective evaluations showed an increase in speech quality for noisy signals enhanced with our method.

The consensus number concept is used to determine the power of synchronization primitives in distributed systems. Recent work in the blockchain domain motivates shifting the attention to consensus number 1 objects, as it has been shown that transaction-based blockchains just need consensus number 1. In this paper we want to get a better understanding of such consensus number 1 objects. In particular, we study the necessary and sufficient conditions for determining the consensus number 1 objects. If an object has consensus number 1, then its operations must be either commutative or associative (necessary condition). On the other hand, if the operations are consistently commutative or overwriting, i.e., independent of the current state of the object, then the consensus number of the object is 1 (sufficient condition). We give an algorithm to implement such generic consensus number 1 objects using only read/write registers. This implies that read/write registers are universal enough to solve tasks, such as asset transfer of a cryptocurrency, among many others, in wait-free distributed systems for any number of processes.

Many deep learning methods have been proposed recently to learn algorithms for combinatorial problems. However, most approaches focus on either supervised/imitation learning (the target algorithm is known) or single agent reinforcement learning (the input distribution is fixed). In some cases, however, the input distribution scales combinatorially as well and cannot easily be fully represented in a concise data set. In this paper, we propose a self-play approach to learn a distributed directory protocol to coordinate concurrent requests to a shared mobile resource among a network of nodes. The self-play is between two agents – a request agent, which finds worst case request inputs – and a route agent, which finds an algorithm that works well on the proposed worst case inputs (and consequently, works well on most queries). We show that our self-play approach is successful in learning an algorithm that works well across diverse request sequences. The empirical performance of the learnt algorithms is within the best known theoretical bounds, and sometimes significantly better than the best known upper bounds.

In this work we provide new insights into the transformer architecture, and in particular, its best-known variant, BERT. First, we propose a method to measure the degree of non-linearity of different elements of transformers. Next, we focus our investigation on the feed-forward networks (FFN) inside transformers, which contain two thirds of the model parameters and have so far not received much attention. We find that FFNs are an inefficient yet important architectural element and that they cannot simply be replaced by attention blocks without a degradation in performance. Moreover, we study the interactions between layers in BERT and show that, while the layers exhibit some hierarchical structure, they extract features in a fuzzy manner. Our results suggest that BERT has an inductive bias towards layer commutativity, which we find is mainly due to the skip connections. This provides a justification for the strong performance of recurrent and weight-shared transformer models.

We consider financial networks, where banks are connected by contracts such as debts or credit default swaps. We study the clearing problem in these systems: we want to know which banks end up in a default, and what portion of their liabilities can these defaulting banks fulfill. We analyze these networks in a sequential model where banks announce their default one at a time, and the system evolves in a step-by-step manner. We first consider the reversible model of these systems, where banks may return from a default. We show that the stabilization time in this model can heavily depend on the ordering of announcements. However, we also show that there are systems where for any choice of ordering, the process lasts for an exponential number of steps before an eventual stabilization. We also show that finding the ordering with the smallest (or largest) number of banks ending up in default is an NP-hard problem. Furthermore, we prove that defaulting early can be an advantageous strategy for banks in some cases, and in general, finding the best time for a default announcement is NP-hard. Finally, we discuss how changing some properties of this setting affects the stabilization time of the process, and then use these techniques to devise a monotone model of the systems, which ensures that every network stabilizes eventually.

Satellite-based localization systems do not work well indoors. Signals sent by aircraft in the ADS-B protocol however can be used for indoor localization. We propose improvements to the multilateration using ADS-B messages and then combine ADS-B multilateration with satellite-based localization. Even in situations where not enough satellites are available to estimate a position, the addition of ADS-B signals allows for high localization accuracy. We evaluate our improvements to the ADS-B multilateration and our combined method using a smartphone and an affordable RTL-SDR.

A set of mutually distrusting participants that want to agree on a common opinion must solve an instance of a Byzantine agreement problem. These problems have been extensively studied in the literature. However, most of the existing solutions assume that the participants are aware of n — the total number of participants in the system — and f — an upper bound on the number of Byzantine participants. In this paper, we show that most of the fundamental agreement problems can be solved without affecting resiliency even if the participants do not know the values of(possibly changing) n and f. Specifically, we consider a synchronous system where the participants have unique but not necessarily consecutive identifiers, and give Byzantine agreement algorithms for reliable broadcast, approximate agreement, rotor-coordinator, early terminating consensus and total ordering in static and dynamic systems, all with the optimal resiliency of n>3f. Moreover, we show that some synchrony is necessary as an agreement with probabilistic termination is impossible in a semi-synchronous or asynchronous system if the participants are unaware of n and f.

We study the evaluation of graph explanation methods. The state of the art to evaluate explanation methods is to first train a GNN, then generate explanations, and finally compare those explanations with the ground truth. We show five pitfalls that sabotage this pipeline because the GNN does not use the ground-truth edges. Thus, the explanation method cannot detect the ground truth. We propose three novel benchmarks: (i) pattern detection, (ii) community detection, and (iii) handling negative evidence and gradient saturation. In a re-evaluation of state-of-the-art explanation methods, we show paths for improving existing methods and highlight further paths for GNN explanation research.

Similar to satellite-based localization systems, messages sent by aircraft with the ADS-B protocol can be used to estimate the location of a mobile receiver. However, for a robust localization using a least-squares approach, ADS-B messages have to be collected over a long time. We propose a localization method based on matching the received signal with known ADS-B messages from distributed receivers. Our proposed method only requires three seconds of recording. Compared to satellite-based localization methods, this approach also works indoors as the signals sent by aircraft are much stronger.

This paper studies Dropout Graph Neural Networks (DropGNNs), a new approach that aims to overcome the limitations of standard GNN frameworks. In DropGNNs, we execute multiple runs of a GNN on the input graph, with some of the nodes randomly and independently dropped in each of these runs. Then, we combine the results of these runs to obtain the final result. We prove that DropGNNs can distinguish various graph neighborhoods that cannot be separated by message passing GNNs. We derive theoretical bounds for the number of runs required to ensure a reliable distribution of dropouts, and we prove several properties regarding the expressive capabilities and limits of DropGNNs. We experimentally validate our theoretical findings on expressiveness. Furthermore, we show that DropGNNs perform competitively on established GNN benchmarks.

We study financial networks where banks are connected by debt contracts. We consider the operation of debt swapping when two creditor banks decide to exchange an incoming payment obligation, thus leading to a locally different network structure. We say that a swap is positive if it is beneficial for both of the banks involved; we can interpret this notion either with respect to the amount of assets received by the banks, or their exposure to different shocks that might hit the system. We analyze various properties of these swapping operations in financial networks. We first show that there can be no positive swap for any pair of banks in a static financial system, or when a shock hits each bank in the network proportionally. We then study worst-case shock models, when a shock of given size is distributed in the worst possible way for a specific bank. If the goal of banks is to minimize their losses in such a worst-case setting, then a positive swap can indeed exist. We analyze the effects of such a positive swap on other banks of the system, the computational complexity of finding a swap, and special cases where a swap can be found efficiently. Finally, we also present some results for more complex swapping operations when the banks swap multiple contracts, or when more than two banks participate in the swap.

The Byzantine agreement problem is considered to be a core problem in distributed systems. For example, Byzantine agreement is often used to build a blockchain, a totally ordered log of records. Blockchains are asynchronous distributed systems, fault-tolerant against Byzantine nodes.

Generating adversarial examples for natural language is hard, as natural language consists of discrete symbols, and examples are often of variable lengths.In this paper, we propose a geometryinspired attack for generating natural language adversarial examples.Our attack generates adversarial examples by iteratively approximating the decision boundary of Deep Neural Networks (DNNs).Experiments on two datasets with two different models show that our attack fools natural language models with high success rates, while only replacing a few words.Human evaluation shows that adversarial examples generated by our attack are hard for humans to recognize.Further experiments show that adversarial training can improve model robustness against our attack.

We present a precise and robust GPS spoofing mitigation method, which is based on a position likelihood distribution. Compared to existing spoofing mitigation methods, this maximum likelihood method is less prone to selecting the wrong satellite signals when (spoofed) duplicates are received. The presented method operates from GPS signal snapshots as short as one millisecond and detects all likely receiver positions. Even with spoofing signals stronger than the authentic satellite signals, the actual receiver position is still found through iterative dampening of the strongest signals. The spoofing mitigation capability is evaluated on the de facto standard TEXBAT spoofing dataset. We reach median position errors introduced by spoofing below 19 m and keep the maximum error below 222 m for all TEXBAT scenarios. This is six times more accurate than the best previous work, which only detects spoofing attacks, but does not mitigate them.

In this paper, we present tight bounds for the k-server problem with delays in the uniform metric space. The problem is defined on n+k nodes in the uniform metric space which can issue requests over time. These requests can be served directly or with some delay using k servers, by moving a server to the corresponding node with an open request. The task is to find an online algorithm that can serve the requests while minimizing the total moving and delay costs. We first provide a lower bound by showing that the competitive ratio of any deterministic online algorithm cannot be better than (2k+1) in the clairvoyant setting. We will then show that conservative algorithms (without delay) can be equipped with an accumulative delay function such that all such algorithms become (2k+1)-competitive in the non-clairvoyant setting. Together, the two bounds establish a tight result for both, the clairvoyant and the non-clairvoyant settings.

A set of participants that want to agree on a common opinion despite the presence of malicious or Byzantine participants need to solve an instance of a Byzantine agreement problem. This classic problem has been well studied but most of the existing solutions assume that the participants are aware of n --- the total number of participants in the system --- and f --- the upper bound on the number of Byzantine participants. In this paper, we examine a synchronous system with Byzantine faults where the participants are neither aware of n nor f. The participants have unique identifiers, which are not necessarily consecutive. For such a system, we give algorithms for rotor-coordinator and consensus, both with the resiliency of n > 3f, which is also the optimal resiliency for solving consensus when the participants know n and f. Thus, resiliency is unaffected even if the Byzantine participants can lie about n and f.

This paper presents a novel shared memory model that simplifies the analysis of consensus on a Chain and a DAG. In this new model, referred to as the append memory model, nodes are allowed to write new values to the unordered memory, but not to overwrite already existing values. We show that although this model differs from the standard shared memory model with n shared read-write registers, many known results from the shared memory model still hold in the append memory model: It is, for example, impossible to establish consensus on n nodes with one crash failure if the nodes in the system are asynchronous. We also consider the append memory model in a synchronous setting with Byzantine failures. For this case, we show that Byzantine agreement cannot be solved in less than t+1 rounds, where t is the number of Byzantine nodes in the system. Assuming a probabilistic access restriction to the append memory, we compare the Byzantine agreement protocols on the Chain and the DAG. We show that the DAG structure achieves an almost optimal resilience (close to t<n/2) in contrast to the Chain structure that can tolerate less than t

Red-blue pebble games model the computation cost of a two-level memory hierarchy. We present various hardness results in different red-blue pebbling variants, with a focus on the oneshot model. We first study the relationship between previously introduced red-blue pebble models (base, oneshot, nodel). We also analyze a new variant (compcost) to obtain a more realistic model of computation. We then prove that red-blue pebbling is NP-hard in all of these model variants. Furthermore, we show that in the oneshot model, a δ-approximation algorithm for δ<2 is only possible if the unique games conjecture is false. Finally, we show that greedy algorithms are not good candidates for approximation, since they can return significantly worse solutions than the optimum.

Abstract Recently, there has been a growing interest in grid exploration by agents with limited capabilities. We show that the grid cannot be explored by three semi-synchronous finite automata, answering an open question by Emek et al. (Theor Comput Sci 608:255–267, 2015) in the negative. In the setting we consider, time is divided into discrete steps, where in each step, an adversarially selected subset of the agents executes one look–compute–move cycle. The agents operate according to a shared finite automaton, where every agent is allowed to have a distinct initial state. The only means of communication is to sense the states of the agents sharing the same grid cell. The agents are equipped with a global compass and whenever an agent moves, the destination cell of the movement is chosen by the agent’s automaton from the set of neighboring grid cells. In contrast to the four agent protocol by Emek et al., we show that three agents do not suffice for grid exploration.

In the context of second layer payments in Bitcoin, and specifically the Lightning Network, we propose a design for a lightweight watchtower that does not need to store signed justice transactions. We alter the structure of the opening and commitment transactions in Lightning channels to encode justice transactions as part of the commitment transactions. With that, a watchtower just needs to watch for specific cheating commitment transaction IDs on the blockchain and can extract signed justice transactions directly from these commitment transactions that appear on the blockchain. Our construction saves an order of magnitude in storage over existing watchtower designs. In addition, we let the watchtower prove to each channel that it has access to all the data required to do its job, and can therefore be paid-per-update.

Human activity recognition from raw sensor data has enabled modern wearable devices to track and analyze everyday activities. However, when used in real world conditions, the performance of off-the-shelf devices is often insufficient. This paper tackles the problem of swimming style recognition and lap counting using sensor data from a single smartwatch. In total 17 hours of this data was collected from 40 swimmers of diverse backgrounds. The data was then used to train a convolutional neural network to recognize the four main swimming styles, transition periods and lap turns. Our method achieves an F1 score of 97.4% for style recognition and 99.2% for counting laps. To the best of our knowledge, these results are the first to enable accurate automatic swimming recognition in a realistic and completely uncontrolled environment.

A differential acoustic OFDM technique is presented to embed data imperceptibly in existing music. The method allows playing back music containing the data with a speaker without users noticing the embedded data channel. Using a microphone, the data can be recovered from the recording. Experiments with smartphone microphones show that transmission distances of 24 meters are possible, while achieving bit error ratios of less than 10 percent, depending on the environment. Furthermore, we present a user study which shows that many people do not recognize the added data channel in music, even when being informed about the experiment and therefore actively listening for the data transmission. Depending on the source music, data rates of 300 to 400 bits per second are achieved.

In this paper we study the problem of monaural music source separation, where a piece of music is to be separated into its main constituent sources. We propose a simple yet effective deep neural network architecture based on a ResNet autoencoder. We investigate several data augmentation and post-processing methods to improve the separation results and outperform various state of the art monaural source separation methods on the DSD100 and MUSDB18 datasets. Our results suggest that in order to further push the state of the art in monaural music source separation we need more data, better data augmentation methods, as well as more effective post-processing methods; and not necessarily ever more complex neural network architectures.

The consensus number of an object is the maximum number of processes among which binary consensus can be solved using any number of instances of the object and read-write registers. Herlihy [1] showed in his seminal work that if an object has a consensus number of n, then its instances can be used to implement any non-trivial object or data structure that is shared among n processes, so that the implementation is wait-free and linearizable. Thus, an object such as compare-and-set with an infinite consensus number is “advanced” because its instances can be used to implement any non-trivial concurrent object shared among any number of processes. On the other hand, objects such as fetch-and-add or fetch-and-multiply have a consensus number of two and are “elementary”. An important consequence of Herlihy's result was that any number of reasonable elementary objects are provably insufficient to implement an advanced object like compare-and-set. However, Ellen et al. [2] observed recently that real multiprocessors do not compute using objects but using instructions that are applied on memory locations. Using this observation, they show that it is possible to use a couple of elementary instructions on the same memory location to implement an advanced one, and consequently any non-trivial object or data structure. However, the above result is only a possibility and uses a generic universal construction as a black-box, which is not how we implement objects in practice, as the generic construction is quite inefficient with respect to the number of steps taken by a process and the number of shared objects used in the worst case. Instead, the efficient implementations are built upon the widely supported compare-and-set instruction and one cannot conclude from the previous result whether the elementary instructions can also produce equally efficient implementations like compare-and-set does or they are fundamentally limited in this respect. In this paper, we answer this question by giving a wait-free and linearizable implementation of compare-and-set using just two elementary instructions, half-max and max-write. The implementation takes O(1) steps per process and uses O(1) shared objects per process. Thus, any known or unknown compare-and-set based implementation can also be done using only two elementary instructions without any loss in efficiency. An interesting aspect of these elementary instructions is that depending on the underlying system, their throughput in a highly concurrent setting is larger than that of the compare-and-set instructions by a factor proportional to n.

We analyze the stabilization time of minority processes in graphs. A minority process is a dynamically changing coloring, where each node repeatedly changes its color to the color which is least frequent in its neighborhood. First, we present a simple $Ω(n^2)$ stabilization time lower bound in the sequential adversarial model. Our main contribution is a graph construction which proves a $Ω(n^{2-ε})$ stabilization time lower bound for any $ε>0$. This lower bound holds even if the order of nodes is chosen benevolently, not only in the sequential model, but also in any reasonable concurrent model of the process.

Consistency in network updates is a nascent research area, especially in the context of traffic engineering or Software Defined Networks. Various approaches have been proposed and implemented in the problem space of flow migration and congestion, primarily focusing on different flows not breaking the bandwidth capacities of the used links during updates. However, current network update techniques overlook the effect of flows congesting their own path during a network update due to latency on the links. Furthermore, while congestion will be resolved eventually after the network update, the buffers of the affected routers can be filled for a long time period, leading to the following paradox: a flow is moved to a path with less latency, but the latency stays the same! As flows are often migrated because of latency concerns, this is highly undesirable. We show that these effects occur already in a small topology in practice, causing packet loss due to overfull buffers. Furthermore, we prove that finding a lossless flow migration is NP-hard, already for a single (splittable) flow on directed acyclic graphs. Nonetheless, we can relax latency requirements to still obtain lossless flow migration. To this end, we show how to adapt current systems such as SWAN or Dionysus [SIGCOMM'13/'14], also developing our own polynomial time schedule algorithm, and discussing future consistent flow migration technique adaptations.

In this paper we consider the problem of designing a distributed directory service. The two classic directory service protocols are Arrow and Ivy. Arrow performs well if the network is a tree, while Ivy performs well on complete graphs. However, there are graphs for which both Arrow and Ivy yield poor performance. In this paper, we propose a new distributed directory protocol, Arvy. Arvy is a natural extension of both Arrow and Ivy, generalizing both, while keeping their simplicity and strengths. Our main contribution is to prove Arvy's correctness, in asynchronous networks with concurrent requests, for arbitrary topologies. Regarding performance, we show that Arvy achieves constant competitive ratio on rings using constant space per node.

A minority process in a weighted graph is a dynamically changing coloring. Each node repeatedly changes its color in order to minimize the sum of weighted conflicts with its neighbors. We study the number of steps until such a process stabilizes. Our main contribution is an exponential lower bound on stabilization time. We first present a construction showing this bound in the adversarial sequential model, and then we show how to extend the construction to establish the same bound in the benevolent sequential model, as well as in any reasonable concurrent model. Furthermore, we show that the stabilization time of our construction remains exponential even for very strict switching conditions, namely, if a node only changes color when almost all (i.e., any specific fraction) of its neighbors have the same color. Our lower bound works in a wide range of settings, both for node-weighted and edge-weighted graphs, or if we restrict minority processes to the class of sparse graphs.

The ability to use a 2D map to navigate a complex 3D environment is quite remarkable, and even difficult for many humans. Localization and navigation is also an important problem in domains such as robotics, and has recently become a focus of the deep reinforcement learning community. In this paper we teach a reinforcement learning agent to read a map in order to find the shortest way out of a random maze it has never seen before. Our system combines several state-of-the-art methods such as A3C and incorporates novel elements such as a recurrent localization cell. Our agent learns to localize itself based on 3D first person images and an approximate orientation angle. The agent generalizes well to bigger mazes, showing that it learned useful localization and navigation capabilities.

In this paper we investigate how to measure real world performance of Android devices using app start durations. To this end we collect ground truth app start times using an automated mechanical setup. The ground truth start times are highly correlated with the outputs from Android's ActivityManager, which we then use to obtain app start times during normal use on a range of rooted devices. We then predict app start times with supervised learning to detect if device performance has changed over time. We show that training data can be gathered on a small set of rooted devices and then applied to other, non rooted devices. We also present an unsupervised method that can track the evolution of the system performance without requiring root access at all.

This paper presents an efficient traffic scheduling algorithm for vehicles such as cars, trains or ships. We provide guarantees for deadlock and starvation freedom, therefore ensuring progress for each vehicle in the system. Our method tolerates vehicles which do not disappear from the traffic network once they reach their destination, but rather continue towards subsequent destinations. Therefore, vehicles can run indefinitely. We introduce the concept of "safe spots", which are locations where a vehicle can stop without ever blocking another vehicle. Using such safe spots, we divide routes into short segments, which reduces the number of routing alternatives exponentially, thus allowing real-time traffic allocation.

Learning in sparse reward settings remains a challenge in Reinforcement Learning, which is often addressed by using intrinsic rewards. One promising strategy is inspired by human curiosity, requiring the agent to learn to predict the future. In this paper a curiosity-driven agent is extended to use these predictions directly for training. To achieve this, the agent predicts the value function of the next state at any point in time. Subsequently, the consistency of this prediction with the current value function is measured, which is then used as a regularization term in the loss function of the algorithm. Experiments were made on grid-world environments as well as on a 3D navigation task, both with sparse rewards. In the first case the extended agent is able to learn significantly faster than the baselines.

Deep generative models such as Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) have recently been applied to style and domain transfer for images, and in the case of VAEs, music. GAN-based models employing several generators and some form of cycle consistency loss have been among the most successful for image domain transfer. In this paper we apply such a model to symbolic music and show the feasibility of our approach for music genre transfer. Evaluations using separate genre classifiers show that the style transfer works well. In order to improve the fidelity of the transformed music, we add additional discriminators that cause the generators to keep the structure of the original music mostly intact, while still achieving strong genre transfer. Visual and audible results further show the potential of our approach. To the best of our knowledge, this paper represents the first application of GANs to symbolic music domain transfer.

We present a generally applicable tree-like sparse multilayer architecture that has a balanced connection from all input neurons to all output neurons. If the ratio between input and output neurons is fixed, the parameters required by our architecture scale with O(n1.5) as compared to O(n2) in a fully connected layer, where n is the number of input neurons. Our sparse 2-layer architecture performs similar and/or superior when compared to its fully connected 1-layer and 2-layer counter parts on the IMDB review sentiment classification task, the Reuters news categorization task and the CIFAR-10 image classification task.

We consider the problem of online Min-cost Perfect Matching with Delays (MPMD) recently introduced by Emek et al, (STOC 2016). This problem is defined on an underlying $n$-point metric space. An adversary presents real-time requests online at points of the metric space, and the algorithm is required to match them, possibly after keeping them waiting for some time. The cost incurred is the sum of the distances between matched pairs of points (the connection cost), and the sum of the waiting times of the requests (the delay cost). We present an algorithm with a competitive ratio of $O(\log n)$, which improves the upper bound of $O(\log^2n+\logΔ)$ of Emek et al, by removing the dependence on $Δ$, the aspect ratio of the metric space (which can be unbounded as a function of $n$). The core of our algorithm is a deterministic algorithm for MPMD on metrics induced by edge-weighted trees of height $h$, whose cost is guaranteed to be at most $O(1)$ times the connection cost plus $O(h)$ times the delay cost of every feasible solution. The reduction from MPMD on arbitrary metrics to MPMD on trees is achieved using the result on embedding $n$-point metric spaces into distributions over weighted hierarchically separated trees of height $O(\log n)$, with distortion $O(\log n)$. We also prove a lower bound of $Ω(\sqrt{\log n})$ on the competitive ratio of any randomized algorithm. This is the first lower bound which increases with $n$, and is attained on the metric of $n$ equally spaced points on a line. The problem of Min-cost Bipartite Perfect Matching with Delays (MBPMD) is the same as MPMD except that every request is either positive or negative, and requests can be matched only if they have opposite polarity. We prove an upper bound of $O(\log n)$ and a lower bound of $Ω(\log^{1/3}n)$ on the competitive ratio of MBPMD with a more involved analysis.

To solve Byzantine agreement, n nodes with real input values, among which t <; n/3 are Byzantine, have to agree on a common consensus value. Previous research has mainly focused on determining a consensus value equal to an input value of some arbitrary node. In this work we instead assume that the values of the nodes are ordered and introduce a novel validity condition which accepts consensus values that are close to the k-th smallest value of the correct nodes. We propose a deterministic algorithm that approximates the k-th smallest value and show that this approximation is the best possible for the synchronous message passing model. Our approach is furthermore extended to multiple dimensions, where the order is not well-defined, and we show that our algorithm can be applied to determine a value that lies within a box around all correct input vectors.

Recently, there has been a growing interest in grid exploration by agents with limited capabilities. We show that the grid cannot be explored by three semi-synchronous finite automata, answering an open question by Emek et al. [TCS'15] in the negative. In the setting we consider, time is divided into discrete steps, where in each step, an adversarially selected subset of the agents executes one look-compute-move cycle. The agents operate according to a shared finite automaton, where every agent is allowed to have a distinct initial state. The only means of communication is to sense the states of the agents sharing the same grid cell. The agents are equipped with a global compass and whenever an agent moves, the destination cell of the movement is chosen by the agent's automaton from the set of neighboring grid cells. In contrast to the four agent protocol by Emek et al., we show that three agents do not suffice for grid exploration.

In the min-cost bipartite perfect matching with delays (MBPMD) problem, requests arrive online at points of a finite metric space. Each request is either positive or negative and has to be matched to a request of opposite polarity. As opposed to traditional online matching problems, the algorithm does not have to serve requests as they arrive, and may choose to match them later at a cost. Our objective is to minimize the sum of the distances between matched pairs of requests (the connection cost) and the sum of the waiting times of the requests (the delay cost). This objective exhibits a natural tradeoff between minimizing the distances and the cost of waiting for better matches. This tradeoff appears in many real-life scenarios, notably, ride-sharing platforms. MBPMD is related to its non-bipartite variant, min-cost perfect matching with delays (MPMD), in which each request can be matched to any other request. MPMD was introduced by Emek et al. (STOC'16), who showed an O(log^2(n)+log(Delta))-competitive randomized algorithm on n-point metric spaces with aspect ratio Delta. Our contribution is threefold. First, we present a new lower bound construction for MPMD and MBPMD. We get a lower bound of Omega(sqrt(log(n)/log(log(n)))) on the competitive ratio of any randomized algorithm for MBPMD. For MPMD, we improve the lower bound from Omega(sqrt(log(n))) (shown by Azar et al., SODA'17) to Omega(log(n)/log(log(n))), thus, almost matching their upper bound of O(log(n)). Second, we adapt the algorithm of Emek et al. to the bipartite case, and provide a simplified analysis that improves the competitive ratio to O(log(n)). The key ingredient of the algorithm is an O(h)-competitive randomized algorithm for MBPMD on weighted trees of height h. Third, we provide an O(h)-competitive deterministic algorithm for MBPMD on weighted trees of height h. This algorithm is obtained by adapting the algorithm for MPMD by Azar et al. to the apparently more complicated bipartite setting.

In this paper we present Disca, a tool to analyze discussions in terms of which person is speaking at what time. We rely on a set of smartphones collaborating in detecting the most likely speaker at every given moment in real time. Each pair of smartphones observes a time difference of arrival pattern that is caused by the location of the different participants. The set of observations between all pairs of smartphones is then used to identify speakers on-line. To achieve this, clock differences and clock drifts between devices are estimated and compensated. Ultimately, participants are found by clustering time difference of arrival measurements which are unique for distinct speakers. We implement the system as an Android application and show that for more than 90% of time windows the correct speaker can be identified. To cope with heterogeneous hardware of Android smartphones, the computational burden is dynamically distributed among all participating smartphones according to their performance.

For the game of Cops and Robbers, it is known that in 1-cop-win graphs, the cop can capture the robber in O(n) time, and that there exist graphs in which this capture time is tight. When k >= 2, a simple counting argument shows that in k-cop-win graphs, the capture time is at most O(n^{k + 1}), however, no non-trivial lower bounds were previously known; indeed, in their 2011 book, Bonato and Nowakowski ask whether this upper bound can be improved. In this paper, the question of Bonato and Nowakowski is answered on the negative, proving that the O(n^{k + 1}) bound is asymptotically tight for any constant k >= 2. This yields a surprising gap in the capture time complexities between the 1-cop and the 2-cop cases.

We propose a novel approach for the generation of polyphonic music based on LSTMs. We generate music in two steps. First, a chord LSTM predicts a chord progression based on a chord embedding. A second LSTM then generates polyphonic music from the predicted chord progression. The generated music sounds pleasing and harmonic, with only few dissonant notes. It has clear long-term structure that is similar to what a musician would play during a jam session. We show that our approach is sensible from a music theory perspective by evaluating the learned chord embeddings. Surprisingly, our simple model managed to extract the circle of fifths, an important tool in music theory, from the dataset.

GPS is used for outdoor localization in a large variety of applications. Current receivers consume too much power for energy-constrained situations like continuous location tracking on small wearable devices. Mainly, this is due to the large amount of GPS signal that has to be decoded to compute the first position fix. While Coarse-Time Navigation (CTN) can reduce the necessary signal to a few milliseconds, it is not robust to noise. Collective Detection (CD) of satellites can mitigate noise to some degree, but the basic method is computationally expensive. We show how CD can be solved optimally and efficiently. Furthermore, we improve the accuracy of CD by exploiting the shape of the likelihood function. All our results are based on real-world signal observations and we achieve localization accuracies of less than 25 meters using a single millisecond of signal. When using 10 consecutive millisecond samples the accuracy improves to less than 10 meters.

We present a new fault-tolerant distributed state machine to inherit the best features of its “parents in spirit”: Paxos, providing strong consistency, and a blockchain, providing simplicity and availability. Our proposal is simple as it does not include any heavy weight distributed failure handling protocols such as leader election. In addition, our proposal has a few other valuable features, e.g., it is responsive, it scales well, and it does not send any overhead messages.

We consider the problem of building a wait-free and linearizable counter using shared registers. The counter supports a read operation, which returns the value of the counter, and an increment operation, which increments the value of the counter and returns nothing. The shared registers support read, write and compare-and-swap instructions. We show that given (n) processes and (O(n)) shared registers, the increment operation is in (O(log n)) and read operation is in (O(1)).

The standard method for outdoor localization is GPS, because it is globally available, relatively accurate and receivers are inexpensive. However, GPS does not work well indoors due to low signal strength.

Contrary to common belief, a recent work by Ellen, Gelashvili, Shavit, and Zhu has shown that computability does not require multicore architectures to support "strong" synchronization instructions like compare-and-swap, as opposed to combinations of "weaker" instructions like decrement and multiply. However, this is the status quo, and in turn, most efficient concurrent data-structures heavily rely on compare-and-swap (e.g. for swinging pointers). We show that this need not be the case, by designing and implementing a concurrent linearizable Log data-structure (also known as a History object), supporting two operations: append(item), which appends the item to the log, and get-log(), which returns the appended items so far, in order. Readers are wait-free and writers are lock-free, hence this data-structure can be used in a lock-free universal construction to implement any concurrent object with a given sequential specification. Our implementation uses atomic read, xor, decrement, and fetch-and-increment instructions supported on X86 architectures, and provides similar performance to a compare-and-swap-based solution on today's hardware. This raises a fundamental question about minimal set of synchronization instructions that the architectures have to support.

We study a fundamental measure for wireless interference in the signal-to-interference noise ratio model known as (weighted) inductive independence. This measure characterizes the effectiveness of using oblivious power---when the power used by a transmitter only depends on the distance to the receiver---as a mechanism for improving wireless capacity. We prove optimal bounds for inductive independence, implying a number of algorithmic applications. An algorithm is provided that achieves capacity that is---due to existing lower bounds---asymptotically best possible using oblivious power assignments. Improved approximation algorithms are provided for a number of problems involving both oblivious power and arbitrary power control, including connectivity, secondary spectrum auctions, and dynamic packet scheduling. We also show that the price of oblivious power---the relative increase in capacity possible when using unconstrained power control---is only doubly logarithmic in the maximum link length.

Traditionally, achieving constructive interference (CI) required specialized timekeeping hardware. Recently, the ability and interest to employ CI distributedly at any time using groups of ordinary single antenna wireless sensor nodes have grown. In this paper, we investigate achieving CI on sensor nodes. We consider the commonly employed IEEE 802.15.4 wireless standard, which uses a chip frequency of 1 MHz. This means signals need to be synchronized with an error below 0.5 microseconds to allow for CI. Hence, excellent clock synchronization between nodes as well as precise transmission timing are required. We implemented and tested a prototype addressing the implementation challenges of synchronizing the nodes' clocks up to a precision of a few hundred nanoseconds and of timing transmissions as accurately as possible. Our results show that, even after multiple minutes of sleep, our approach is able to achieve CI in over 30% of cases, in scenarios in which any influence from the capture effect can be ruled out. This leads to an increase in a packet's chance of arrival to 30-65%, compared to 0-30% when transmitting with either less synchrony or different data payload. Further, we find that 2 senders generally increase the signal power by 2-3 dB and can double the packet reception ratio of weak links.

We study complexity and algorithms for network updates in the setting of Software Defined Networks. Our focus lies on consistent updates for the case of updating forwarding rules in a loop free manner and the migration of flows without congestion. In both cases, we study how the power of two affects the respective problem setting. For loop freedom, we show that scheduling consistent updates for two destinations is NP-hard for a sublinear number of rounds. We also consider the dynamic case, and show that this problem is NP-hard as well via a reduction from Feedback Arc Set. While the power of two increases the complexity for loop freedom, the converse is true when allowing to split flows twice. For the NP-hard problem of consistently migrating unsplittable flows to new routes while respecting waypointing and service chains, we prove that two-splittability allows the problem to be tractable again.

We study consistent migration of flows, with special focus on software defined networks. Given a current and a desired network flow configuration, we give the first polynomial-time algorithm to decide if a congestion-free migration is possible. However, if all flows must be integer or are unsplittable, this is NP-hard to decide. A similar problem is providing increased bandwidth to an application, while keeping all other flows in the network, but possibly migrating them consistently to other paths. We show that the maximum increase can be approximated arbitrarily well in polynomial time. Current methods as RSVP-TE consider unsplittable flows and remove flows of lesser importance in order to increase bandwidth for an application: We prove that deciding what flows need to be removed is an NP-hard optimization problem with no PTAS possible unless P = NP.

We present RTDS, an Android application to analyze discussions while they are taking place. Using two microphones of a smart phone and Time Difference of Arrival measurements, conversations of participants are evaluated regarding, e.g., speaking time, contributions, or complex interaction patterns. The application can also assume the role of an active referee to ensure that all speakers get a fair share of the on-going conversation. By using an off the shelf smart phone with two microphones, our system can immediately be applied to track spoken interactions between people. Experimental results show that our implementation causes only 2% user classification errors whilst being able to run in real-time on a standard smart phone without hardware modifications.

TCP handles packet loss in the network by retransmitting lost packets, which in turn increases latency. Many connections in data centers are short-lived and consist only of a few packets (e.g., RPCs). Such connections suffer disproportionately from packet retransmissions. We address this issue by introducing a new transport layer protocol called ATP: ATP uses ample forward error correction at the beginning of a connection, allowing short-lived flows to recover from packet loss without retransmissions - but at the same time not congesting long-lived flows. Our experiments show that in an environment with background traffic, the latency's 99th percentile can be reduced by a factor of almost 20 while being fair to other TCP connections.

Consider a complete bipartite graph of 2n nodes with n nodes on each side. In a round, each node can either send at most one message to a neighbor or receive at most one message from a neighbor. Each node has a preference list that ranks all its neighbors in a strict order from 1 to n. We introduce a non-negative similarity parameter D < n for the preference lists of nodes on one side only. For D = 0, these preference lists are same and for D = n-1, they can be completely arbitrary. There is no restriction on the preference lists of the other side. We show that each node can compute its partner in a stable matching by receiving O(n(D + 1)) messages of size O(log n) each. We also show that this is optimal (up to a logarithmic factor) if D is constant.

We propose a new class of wireless transmission schemes decoupling synchronization headers from payloads to create new transmission primitives involving a second sender. By transmitting a synchronization header only we can let nearby nodes receive fragments of a packet without having to receive that packet's synchronization header, and by using the capture effect we can overwrite portions of the payload of longer ongoing packets.

This paper studies a new online problem, referred to as min-cost perfect matching with delays (MPMD), defined over a finite metric space (i.e., a complete graph with positive edge weights obeying the triangle inequality) M that is known to the algorithm in advance. Requests arrive in a continuous time online fashion at the points of M and should be served by matching them to each other. The algorithm is allowed to delay its request matching commitments, but this does not come for free: the total cost of the algorithm is the sum of metric distances between matched requests plus the sum of times each request waited since it arrived until it was matched. A randomized online MPMD algorithm is presented whose competitive ratio is O (log2 n + logΔ), where n is the number of points in M and Δ is its aspect ratio. The analysis is based on a machinery developed in the context of a new stochastic process that can be viewed as two interleaved Poisson processes; surprisingly, this new process captures precisely the behavior of our algorithm. A related problem in which the algorithm is allowed to clear any unmatched request at a fixed penalty is also addressed. It is suggested that the MPMD problem is merely the tip of the iceberg for a general framework of online problems with delayed service that captures many more natural problems.

The question of what can be computed, and how efficiently, is at the core of computer science. Not surprisingly, in distributed systems and networking research, an equally fundamental question is what can be computed in a distributed fashion. More precisely, if nodes of a network must base their decision on information in their local neighborhood only, how well can they compute or approximate a global (optimization) problem? In this paper we give the first polylogarithmic lower bound on such local computation for (optimization) problems including minimum vertex cover, minimum (connected) dominating set, maximum matching, maximal independent set, and maximal matching. In addition, we present a new distributed algorithm for solving general covering and packing linear programs. For some problems this algorithm is tight with the lower bounds, whereas for others it is a distributed approximation scheme. Together, our lower and upper bounds establish the local computability and approximability of a large class of problems, characterizing how much local information is required to solve these tasks.

Decentralized monitoring and alarming systems can be an attractive alternative to centralized architectures. Distributed sensor nodes (e.g., in the smart grid's distribution network) are closer to an observed event than a global and remote observer or controller. This improves the visibility and response time of the system. Moreover, in a distributed system, local problems may also be handled locally and without overloading the communication network. This paper studies alarming from a distributed computing perspective and for two fundamentally different scenarios: on-duty and off-duty. We model the alarming system as a sensor network consisting of a set of distributed nodes performing local measurements to sense events. In order to avoid false alarms, the sensor nodes cooperate and only escalate an event (i.e., raise an alarm) if the number of sensor nodes sensing an event exceeds a certain threshold. In the on-duty scenario, nodes not affected by the event can actively help in the communication process, while in the off-duty scenario, non-event nodes are inactive. We present and analyze algorithms that minimize the reaction time of the monitoring system while avoiding unnecessary message transmissions. We investigate time and message complexity tradeoffs in different settings, and also shed light on the optimality of our algorithms by deriving cost lower bounds for distributed alarming systems.

In cloud gaming, servers perform remote rendering on behalf of thin clients. Such a server must deliver sufficient frame rate (at least 30fps) to each of its clients. At the same time, each client desires an immersive experience, and therefore the server should also provide the best graphics quality possible to each client. Statically provisioning time slices of the server GPU for each client suffers from severe underutilization because clients can come and go, and scenes that the clients need rendered can vary greatly in terms of GPU resource usage over time.

Consider a group of mobile finite automata, referred to as agents, located in the origin of an infinite grid. The grid is occupied by obstacles, i.e., sets of cells that can not be entered by the agents. In every step, an agent can sense the states of the co-located agents and is allowed to move to any neighboring cell of the grid not blocked by an obstacle. We assume that the circumference of each obstacle is finite but allow the number of obstacles to be unbounded. The task of the agents is to cooperatively find a treasure, hidden in the grid by an adversary. In this work, we show how the agents can utilize their simple means of communication and their constant memory to systematically explore the grid and to locate the treasure in finite time. As integral part of the agents' behavior, we present a method that allows a group of six agents to follow a straight line, even if the line is partially obstructed by obstacles, and to discover all free cells along this line. In total, our search protocol requires nine agents.

We introduce a stronger validity property for the byzantine agreement problem with orderable initial values: The median validity property. In particular, the decision value is required to be close to the median of the initial values of the non-byzantine nodes. The proximity to the median scales with the desired level of fault-tolerance: If no fault-tolerance is required, algorithms have to decide for the true median. If the number of failures is maximal, algorithms must still decide on a value within the range of the input values of the non-byzantine nodes. We present a deterministic algorithm satisfying this property for n >= 3t+1 within t+1 phases, where t is the maximum number of byzantine nodes and n is the total number of nodes.

The Internet has developed into the primary means of communication, while ensuring availability and stability is becoming an increasingly challenging task. Traffic monitoring enables network operators to comprehend the composition of traffic flowing through individual corporate and private networks, making it essential for planning, reporting and debugging purposes. Classical packet capture and aggregation concepts (e.g. NetFlow) typically rely on centralized collection of traffic metadata. With the proliferation of network enabled devices and the resulting increase in data volume, such approaches suffer from scalability issues, often prohibiting the transfer of raw metadata as such. This paper describes a decentralized approach, eliminating the need for a central collector and storing local views of network traffic patterns on the respective devices performing the capture. In order to allow for the analysis of captured data, queries formulated by analysts are distributed across all devices. Processing takes place in a parallelized fashion on the respective local data. Consequently, instead of continually transferring raw metadata, significantly smaller aggregate results are sent to a central location which are then combined into the requested final result. The proposed system describes a lightweight and scalable monitoring solution, enabling the efficient use of available system resources on the distributed devices, hence allowing for high performance, real-time traffic analysis on a global scale. The solution was implemented and deployed globally on hosts managed and maintained by a large managed network security services provider.

In this paper we consider the trade-off between space and write overhead of flash memory. Every flash memory has additional space to compensate for wear leveling; we denote the space overhead with σ. Furthermore, every flash memory is forced to rewrite valid data when a block is erased; we denote the write overhead with ω. We show that space and write overhead are inversely proportional with σω ≥ 1. We also present an algorithm that proves that our analysis is tight, as it achieves σω = 1 in a worst case. Moreover, we analyze a setting with the data being updated uniformly at random, or not at all.

Clock synchronization is an enabling service for a wide range of applications and protocols in both wired and wireless networks. We study the implications of clock drift and communication latency on the accuracy of clock synchronization when scaling the network diameter. Starting with a theoretical analysis of synchronization protocols, we prove tight bounds on the synchronization error in a model that assumes independently and randomly distributed communication delays and slowly changing drifts. While this model is more optimistic than traditional worst-case analysis, it much better captures the nature of real-world systems such as wireless networks. The bound on the synchronization accuracy, which is roughly the square root of the network diameter, is achieved by the novel PulseSync protocol. Extensive experiments demonstrate that PulseSync is able to meet the predictions from theory and tightly synchronizes large networks. This contrasts against an exponential growth of the skew incurred by the state-of-the-art protocol for wireless sensor networks. Moreover, PulseSync adapts much faster to network dynamics and changing clock drifts than this protocol.

Researchers in ad hoc networks are used to study some of the prevalent technical difficulties in networking: Without a fixed infrastructure, challenges such as mobility or wireless communication must be investigated in their purest form. As a consequence one learns a great deal about the fundamentals of networking. In my talk I will present a few exciting research questions that I learned when studying ad hoc networks: I will discuss how mobility leads to distributed complexity, why clock synchronization is unexpectedly difficult, and what difference it makes to assume a more accurate wireless model. Along my talk I will present several open questions.

Using mobile phones while walking for activities that require continuous focus on the screen, such as texting, has become more and more popular in the last years. To avoid colliding with obstacles, such as lampposts and pedestrians, focus has to be taken off the screen in regular intervals. In this paper we introduce SpareEye, an Android application that warns the smartphone user from obstacles in her way. We use only the camera of the phone and no special hardware, ensuring that it requires minimal effort from the user to use the application during everyday life. Experimental results show that we can detect obstacles with high accuracy, with only some false positives and few false negatives.

This paper considers the computational power of anonymous message passing algorithms (henceforth, anonymous algorithms), i.e., distributed algorithms operating in a network of unidentified nodes. We prove that every problem that can be solved (and verified) by a randomized anonymous algorithm can also be solved by a deterministic anonymous algorithm provided that the latter is equipped with a 2-hop coloring of the input graph. Since the problem of 2-hop coloring a given graph (i.e., ensuring that two nodes with distance at most 2 have different colors) can by itself be solved by a randomized anonymous algorithm, it follows that with the exception of a few mock cases, the execution of every randomized anonymous algorithm can be decoupled into a generic preprocessing randomized stage that computes a 2-hop coloring, followed by a problem-specific deterministic stage. The main ingredient of our proof is a novel simulation method that relies on some surprising connections between 2-hop colorings and an extensively used graph lifting technique.

We present Dionysus, a system for fast, consistent network updates in software-defined networks. Dionysus encodes as a graph the consistency-related dependencies among updates at individual switches, and it then dynamically schedules these updates based on runtime differences in the update speeds of different switches. This dynamic scheduling is the key to its speed; prior update methods are slow because they pre-determine a schedule, which does not adapt to runtime conditions. Testbed experiments and data-driven simulations show that Dionysus improves the median update speed by 53--88% in both wide area and data center networks compared to prior methods.

In this paper, we address two basic questions in wireless communication. First, how long does it take to schedule an arbitrary set of communication requests? Second, given a set of communication requests, how many of them can be scheduled concurrently? Our results are derived in the signal-to-interference-plus-noise ratio (SINR) interference model with geometric path loss and consist of efficient algorithms that find a constant approximation for the second problem and a logarithmic approximation for the first problem. In addition, we show that the interference model is robust to various factors that can influence the signal attenuation. More specifically, we prove that as long as influences on the signal attenuation are constant, they affect the capacity only by a constant factor.

We argue for the development of efficient methods to update the data plane state of an SDN, while maintaining desired consistency properties (e.g., no packet should be dropped). We highlight the inherent trade-off between the strength of the consistency property and dependencies it imposes among rules at different switches; these dependencies fundamentally limit how quickly data plane can be updated. For one basic consistency property---no packet should loop---we develop an update algorithm that has provably minimal dependency structure. We also sketch a general architecture for consistent updates that separates the twin concerns of consistency and efficiency.

Cashless payments are nowadays ubiquitous and decentralized digital currencies like Bitcoin are increasingly used as means of payment. However, due to the delay of the transaction confirmation in Bitcoin, it is not used for payments that rely on quick transaction confirmation. We present a concept that addresses this drawback of Bitcoin and allows it to be used for fast transactions. We evaluate the performance of the concept using double-spending attacks and show that, employing our concept, the success of such attacks diminishes to less than 0.09%. Moreover, we present a real world application: We modified a snack vending machine to accept Bitcoin payments and make use of fast transaction confirmations.

BitTorrent, the most popular peer-to-peer (P2P) file-sharing protocol, accounts for a significant fraction of the traffic of the Internet. Using a novel technique, we measure live BitTorrent swarms on the Internet and confirm the conjecture that overlay networks formed by BitTorrent are not locality-aware, i.e., they include many unnecessary long distance connections. Attempts to improve the locality have failed because they require a modification of the existing protocol, or interventions by Internet service providers (ISPs). In contrast, we propose a lightweight method that improves the locality of active swarms by 6% by suggesting geographically close peers with the Peer Exchange Protocol (PEX), without any modifications to the current system. An improvement of locality not only benefits the ISPs by reducing network transit cost, it also reduces the traffic over long-distance connections, which delays the need to expand the infrastructure, easing the power consumption. We expect that if used on a large scale our method reduces the Internet's energy consumption by 8 TWh a year.

Bitcoin is a digital currency that unlike traditional currencies does not rely on a centralized authority. Instead Bitcoin relies on a network of volunteers that collectively implement a replicated ledger and verify transactions. In this paper we analyze how Bitcoin uses a multi-hop broadcast to propagate transactions and blocks through the network to update the ledger replicas. We then use the gathered information to verify the conjecture that the propagation delay in the network is the primary cause for blockchain forks. Blockchain forks should be avoided as they are symptomatic for inconsistencies among the replicas in the network. We then show what can be achieved by pushing the current protocol to its limit with unilateral changes to the client's behavior.

A new model that depicts a network of randomized finite state machines operating in an asynchronous environment is introduced. This model, that can be viewed as a hybrid of the message passing model and cellular automata is suitable for applying the distributed computing lens to the study of networks of sub-microprocessor devices, e.g., biological cellular networks and man-made nano-networks.

We present SWAN, a system that boosts the utilization of inter-datacenter networks by centrally controlling when and how much traffic each service sends and frequently re-configuring the network's data plane to match current traffic demand. But done simplistically, these re-configurations can also cause severe, transient congestion because different switches may apply updates at different times. We develop a novel technique that leverages a small amount of scratch capacity on links to apply updates in a provably congestion-free manner, without making any assumptions about the order and timing of updates at individual switches. Further, to scale to large networks in the face of limited forwarding table capacity, SWAN greedily selects a small set of entries that can best satisfy current demand. It updates this set without disrupting traffic by leveraging a small amount of scratch capacity in forwarding tables. Experiments using a testbed prototype and data-driven simulations of two production networks show that SWAN carries 60% more traffic than the current practice.

Datacenter networks (DCNs) are constantly evolving due to various updates such as switch upgrades and VM migrations. Each update must be carefully planned and executed in order to avoid disrupting many of the mission-critical, interactive applications hosted in DCNs. The key challenge arises from the inherent difficulty in synchronizing the changes to many devices, which may result in unforeseen transient link load spikes or even congestions. We present one primitive, zUpdate, to perform congestion-free network updates under asynchronous switch and traffic matrix changes. We formulate the update problem using a network model and apply our model to a variety of representative update scenarios in DCNs. We develop novel techniques to handle several practical challenges in realizing zUpdate as well as implement the zUpdate prototype on OpenFlow switches and deploy it on a testbed that resembles real DCN topology. Our results, from both real-world experiments and large-scale trace-driven simulations, show that zUpdate can effectively perform congestion-free updates in production DCNs.

We study a fundamental measure for wireless interference in the SINR model known as (weighted) inductive independence. This measure characterizes the effectiveness of using oblivious power — when the power used by a transmitter only depends on the distance to the receiver — as a mechanism for improving wireless capacity. We prove optimal bounds for inductive independence, implying a number of algorithmic applications. An algorithm is provided that achieves — due to existing lower bounds — capacity that is asymptotically best possible using oblivious power assignments. Improved approximation algorithms are provided for a number of problems for oblivious power and for power control, including distributed scheduling, connectivity, secondary spectrum auctions, and dynamic packet scheduling.

In this paper we study the problem of scheduling wireless links in a model where successive interference cancellation is combined with the traditional physical interference model. Successive interference cancellation is based on the observation that interfering signals should not be treated as random noise, but as well-structured signals. By exploiting this structured nature, the strongest signal can be decoded and subtracted from a collision, thus enabling the decoding of weaker simultaneous signals. The procedure can be repeated iteratively as long as the collided signals differ in strength significantly. It has been shown that the problem of scheduling wireless links with successive interference cancellation is NP-hard. In this work, we propose a polynomial-time scheduling algorithm that uses successive interference cancellation to compute short schedules for network topologies formed by nodes arbitrarily distributed in the Euclidean plane. We prove that the proposed algorithm is correct in the physical interference model and provide simulation results demonstrating the performance of the algorithm in different network topologies. We compare the results to solutions without successive interference cancellation and observe that throughput gains of up to 20% are obtained in certain scenarios.

Tit-for-tat trading lies at the heart of many incentive mechanisms for distributed systems where participants are anonymous. However, since the standard tit-for-tat approach is restricted to bilateral exchanges, data is transferred only between peers with direct and mutual interests. Generalizing tit-for-tat to multi-lateral trades where contributions can occur along cycles of interest may improve the performance of a system in terms of faster downloads without compromising the incentive-compatibility inherent to tit-for-tat trading. In this paper, we study the potential benefits and limitations of such a generalized trading in swarm-based peer-to-peer systems. Extensive simulations are performed to evaluate different techniques and to identify the crucial parameters influencing the obtainable throughput improvements and the corresponding tradeoffs. Moreover, we discuss extensions for overhead reduction and provide an optimized distributed implementation of our techniques. In summary, we find that allowing inter-swarm trades on short trading cycles can improve the throughput significantly; on the other hand, trading on long cycles does not pay off as the communication and management overhead becomes exceedingly large while the additional performance gains are marginal.

We present an algorithm to compute All Pairs Shortest Paths (APSP) of a network in a distributed way. The model of distributed computation we consider is the message passing model: in each synchronous round, every node can transmit a different (but short) message to each of its neighbors. We provide an algorithm that computes APSP in O(n) communication rounds, where n denotes the number of nodes in the network. This implies a linear time algorithm for computing the diameter of a network. Due to a lower bound these two algorithms are optimal up to a logarithmic factor. Furthermore, we present a new lower bound for approximating the diameter D of a graph: Being allowed to answer D+1 or D can speed up the computation by at most a factor D. On the positive side, we provide an algorithm that achieves such a speedup of D and computes an (1+εepsilon) multiplicative approximation of the diameter. We extend these algorithms to compute or approximate other problems, such as girth, radius, center and peripheral vertices. At the heart of these approximation algorithms is the S-Shortest Paths problem which we solve in O(|S|+D) time.

We study the problem of computing the diameter of a network in a distributed way. The model of distributed computation we consider is: in each synchronous round, each node can transmit a different (but short) message to each of its neighbors. We provide an lower bound for the number of communication rounds needed, where n denotes the number of nodes in the network. This lower bound is valid even if the diameter of the network is a small constant. We also show that a (3/2 − ε)-approximation of the diameter requires rounds. Furthermore we use our new technique to prove an lower bound on approximating the girth of a graph by a factor 2 − ε.

In this paper, we study the information exchange problem on a set of multiple access channels: k arbitrary nodes have information they want to distribute to the entire network via a shared medium partitioned into channels. We present algorithms and lower bounds on the time and channel complexity for disseminating these k information items in a single-hop network of n nodes. More precisely, we devise a deterministic algorithm running in asymptotically optimal time O(k) using O(n(log (k)/k)) channels if k less or equal to (1/6) * log n and O(log(1+p) (n/k) channels otherwise, where p>0 is an arbitrarily small constant. In addition, we show that Omega(n(Ω(1/k))+logk n) channels are necessary to achieve this time complexity.

We study the verification problem in distributed networks, stated as follows. Let $H$ be a subgraph of a network $G$ where each vertex of $G$ knows which edges incident on it are in $H$. We would like to verify whether $H$ has some properties, e.g., if it is a tree or if it is connected (every node knows at the end of the process whether $H$ has the specified property or not). We would like to perform this verification in a decentralized fashion via a distributed algorithm. The time complexity of verification is measured as the number of rounds of distributed communication. In this paper we initiate a systematic study of distributed verification and give almost tight lower bounds on the running time of distributed verification algorithms for many fundamental problems such as connectivity, spanning connected subgraph, and $s$-$t$ cut verification. We then show applications of these results in deriving strong unconditional time lower bounds on the hardness of distributed approximation for many classical optimization problems including minimum spanning tree (MST), shortest paths, and minimum cut. Many of these results are the first nontrivial lower bounds for both exact and approximate distributed computation, and they resolve previous open questions. Moreover, our unconditional lower bound of approximating MST subsumes and improves upon the previous hardness of approximation bound of Elkin [M. Elkin, SIAM J. Comput., 36 (2006), pp. 433--456] as well as the lower bound for (exact) MST computation of Peleg and Rubinovich [D. Peleg and V. Rubinovich, SIAM J. Comput., 30 (2000), pp. 1427--1442]. Our result implies that there can be no distributed approximation algorithm for MST that is significantly faster than the current exact algorithm for any approximation factor. Our lower bound proofs show an interesting connection between communication complexity and distributed computing which turns out to be useful in establishing the time complexity of exact and approximate distributed computation of many problems.

While the sun is important for our health, overexposure to sunlight carries significant health risks ranging from sunburn to skin cancer. Although people know about these risks, sunlight related skin damages have increased over the past decades. We have conducted a survey that sheds light on this phenomenon and suggests that the missing natural sense for UV radiation negatively influences people's sun related behavior. To address this issue, we have implemented Sundroid. Sundroid measures the incident UV radiation using a body-worn sensing unit that communicates wirelessly with the user's smartphone. The phone thereby acts as a user interface to present the measured data in an intuitive manner, and to notify the user once a critical amount of sunlight has been reached. Sundroid can also be applied in other contexts, such as behavioral research or medicine. We show that after calibration, errors are within 5% compared to a high-precision reference signal.

We consider the question of how a conspiring subgroup of peers in a p2p network can find each other and communicate without provoking suspicion among regular peers or an authority that monitors the network. In particular, we look at the problem of how a conspirer can broadcast a message secretly to all fellow conspirers. As a subproblem of independent interest, we study the problem of how a conspirer can safely determine a connected peer's type, i.e., learning whether the connected peer is a conspirer or a regular peer without giving away its own type in the latter case. For several levels of monitoring, we propose distributed and efficient algorithms that transmit hidden information by varying the block request sequence meaningfully. We find that a p2p protocol offers several steganographic channels through which hidden information can be transmitted, and p2p networks are susceptible to hidden communication even if they are completely monitored.

This article presents an algorithm for detecting and disseminating information in a single-hop multi-channel wireless network: k arbitrary nodes have information they want to share with the entire network. Neither the nodes that have information nor the number k of these nodes are known initially. This communication primitive lies between the two other fundamental primitives regarding information dissemination: broadcasting (one-to-all communication) and gossiping (total information exchange). The time complexity of the algorithm is linear in the number of information items and thus asymptotically optimal with respect to time. The algorithm does not require collision detection and thanks to using several channels the lower bound of Ω(k+log n) established for single-channel communication can be broken.

A maximal independent set on a graph is an inclusion-maximal set of mutually non-adjacent nodes. This basic symmetry breaking structure is vital for many distributed algorithms, which by now has been fueling the search for fast local algorithms to find such sets over several decades. In this paper, we present a solution with randomized running time O(√log n log log n) on trees, improving roughly quadratically on the state-of-the-art bound. Our algorithm is uniform and nodes need to exchange merely O(log n) many bits with high probability. In contrast to previous techniques achieving sublogarithmic running times, our approach does not rely on any bound on the number of independent neighbors (possibly with regard to an orientation of the edges).

The problem of gathering n autonomous robots in the Euclidean plane at one (not predefined) point is well-studied under various restrictions on the capabilities of the robots and in several time models. However, only very few runtime bounds are known. We consider the scenario of local algorithms in which the robots can only observe their environment within a fixed viewing range and have to base their decision where to move in the next step solely on the relative positions of the robots within their viewing range. Such local algorithms have to guarantee that the (initially connected) unit disk graph defined by the viewing range of the robots stays connected at all times. In this paper, we focus on the synchronous setting in which all robots are activated concurrently. Ando et al.

We explore the fundamental limits of distributed balls-into-bins algorithms, i.e., algorithms where balls act in parallel, as separate agents. This problem was introduced by Adler et al., who showed that non-adaptive and symmetric algorithms cannot reliably perform better than a maximum bin load of Theta(log log n / log log log n) within the same number of rounds. We present an adaptive symmetric algorithm that achieves a bin load of two in log* n+O(1) communication rounds using O(n) messages in total. Moreover, larger bin loads can be traded in for smaller time complexities. We prove a matching lower bound of (1-o(1))log* n on the time complexity of symmetric algorithms that guarantee small bin loads at an asymptotically optimal message complexity of O(n). The essential preconditions of the proof are (i) a limit of O(n) on the total number of messages sent by the algorithm and (ii) anonymity of bins, i.e., the port numberings of balls are not globally consistent. In order to show that our technique yields indeed tight bounds, we provide for each assumption an algorithm violating it, in turn achieving a constant maximum bin load in constant time.

We study the verification problem in distributed networks, stated as follows. Let H be a subgraph of a network G where each vertex of G knows which edges incident on it are in H. We would like to verify whether H has some properties, e.g., if it is a tree or if it is connected (every node knows in the end of the process whether H has the specified property or not). We would like to perform this verification in a decentralized fashion via a distributed algorithm. The time complexity of verification is measured as the number of rounds of distributed communication.

This article reports on the results of our measurement study of the Kad network. Although several fully decentralized peer-to-peer systems have been proposed in the literature, most existing systems still employ a centralized architecture. The Kad ne

We advocate a novel programming approach we call slotted programming that not only addresses the specific hardware capabilities of sensor nodes, but also facilitates coding through a truly modular design. The approach is based on the temporal decoupling of the different tasks of a sensor node such that at any time at most one task is active. In contrast to traditional sensor network programming, slotted programming guarantees that each of these tasks can be implemented as an independent software module, simplifying not only the coding and testing phase, but also the code reuse in a different context. In addition, we believe that the proposed approach is highly qualified for energy efficient and real time applications. To substantiate our claims, we have implemented slotos, an extension to TinyOS that supports slotted programming. Within this framework, we demonstrate the advantages of the slotted programming paradigm.

The size of personal music collections has constantly increased over the past years. As a result, the traditional metadata based lists to browse these collections have reached their limits. Interfaces that are based on music similarity offer an alternative and thus are increasingly gaining attention. Music similarity is typically either derived from audio-features (objective approach) or from user driven information sources, such as collaborative filtering or social tags (subjective approach). Studies show that the latter techniques outperform audio-based approaches when it comes to describe the perceived music similarity. However, subjective approaches typically only define pairwise relations as opposed to the global notion of similarity given by audio-feature spaces. Many of the proposed interfaces for similarity based music access inherently depend on this global notion and are thus not applicable to user driven music similarity measures. The first contribution of this paper is a high dimensional music space that is based on user driven similarity measures. It combines the advantages of audio-feature spaces (global view) with the advantages of subjective sources that better reflect the users' perception. The proposed space compactly represents similarity and therefore is well suited for offline use, such as in mobile applications. To demonstrate the practical applicability, the second contribution is a comprehensive mobile music player that incorporates several smart interfaces to access the user's music collection. Based on this application, we finally present a large-scale user study that underlines the benefits of the introduced interfaces and shows their great user acceptance.

Wireless networks often experience a significant amount of churn, the arrival and departure of nodes. We propose a distributed algorithm that detects churn and is resilient to a worst-case adversary. The nodes of the network are notified about changes quickly, in asymptotically optimal time up to an additive logarithmic overhead.

We demonstrate how to leverage a system's capability for all-to-all communication to achieve an exponential speed-up of local algorithms despite bandwidth and memory restrictions. More precisely, if a network comprises n nodes with all-to-all bandwidth nε (ε > 0 constant) and nodes know their input and neighborhood with respect to a graph problem instance of polylogarithmic maximum degree, any local algorithm for this problem with running time r ∈ O(log n) and reasonably small states can be simulated within O(log r) rounds.

We introduce Multi-Trials, a new technique for symmetry breaking for distributed algorithms and apply it to various problems in general graphs. For instance, we present three randomized algorithms for distributed (vertex or edge) coloring improving on previous algorithms and showing a time/color trade-off. To get a Δ+1 coloring takes time O(log Δ+ √ log n). To obtain an O(Δ+log1+1/log*nn) coloring takes time O(log* n). This is more than an exponential improvement in time for graphs of polylogarithmic degree. Our fastest algorithm works in constant time using O(Δlog(c) n+ log1+1/c n) colors, where c denotes an arbitrary constant and log(c ) n denotes the c times (recursively) applied logarithm ton.

We show how to partition data structures representable by directed acyclic graphs, i.e. rooted trees, to allow for efficient complex operations, which lie beyond inserts, deletes and finds. The approach potentially improves the performance of any operation modifying more than one element of the data structure. It covers common data structures implementable via linked lists or trees such as sets and maps. We demonstrate its simplicity and its effectiveness using a concurrent sorted linked list. We achieve a speedup of up to 250% even for small divisions.

We give a graph decomposition technique that creates entirely independent subproblems for graph problems such as coloring and dominating sets that can be solved without synchronization on a distributed memory system. For coloring, evaluation shows a performance gain of a factor 3 to 5 at the price of using more colors.

Data gathering is one of the most common applications of wireless sensor networks. Such networks are an extremely useful tool for researchers in various domains since they allow for measurements in unaccessible locations, e.g., on mountains, glaciers or in animal habitats. At the same time, the deployment and maintenance of such sites is time consuming and costly. Network reprogramming protocols allow to distribute code updates over the wireless network, which prolongs the interval between human intervention at the deployment site.

Having access to accurate position information is a key requirement for many wireless sensor network applications. We present SpiderBat, an ultrasound-based ranging platform for wireless sensor networks. It is designed to extend existing node platforms with ultrasound ranging capabilities. SpiderBat features four pairs of independently controllable ultrasound transmitters and receivers, which point towards different directions of the compass. Having multiple ultrasound transmitters and receivers provides an increased spatial coverage. Furthermore, our design allows to determine the angle of arrival of an ultrasound wave, which can be exploited for improved accuracy of the positioning.

We present a novel clock synchronization algorithm and prove tight upper and lower bounds on the worst-case clock skew that may occur between any two participants in any given distributed system. More importantly, the worst-case clock skew between neighboring nodes is (asymptotically) at most a factor of two larger than the best possible bound. While previous results solely focused on the dependency of the skew bounds on the network diameter, we prove that our techniques are optimal also with respect to the maximum clock drift, the uncertainty in message delays, and the imposed bounds on the clock rates. The presented results all hold in a general model where both the clock drifts and the message delays may vary arbitrarily within pre-specified bounds. Furthermore, our algorithm exhibits a number of other highly desirable properties. First, the algorithm ensures that the clock values remain in an affine linear envelope of real time. A better worst-case bound on the accuracy with respect to real time cannot be achieved in the absence of an external timer. Second, the algorithm minimizes the number and size of messages that need to be exchanged in a given time period. Moreover, only a small number of bits must be stored locally for each neighbor. Finally, our algorithm can easily be adapted for a variety of other prominent synchronization models.

Recent technological advances foster the spreading of social software in the mobile domain. Hence, future usage patterns of mobile devices will involve more group interaction. While collaboration using mobile devices is an active area of research, only limited attention has been paid to the efficient initiation of group communication from mobile terminals. In this paper we present a community-aware mechanism that allows to efficiently select contacts in order to address them as a group. We have integrated the proposed method into a proof-of-concept application, and present preliminary experiments that demonstrate the accuracy of the approach and show significant time savings in the group initialization process.

Greedy routing is a novel routing paradigm where messages are always forwarded to the neighbor that is closest to the destination. Our main result is a polynomial-time algorithm that embeds combinatorial unit disk graphs (CUDGs - a CUDG is a UDG without any geometric information) into O(log2n)- dimensional space, permitting greedy routing with constant stretch. To the best of our knowledge, this is the first greedy embedding with stretch guarantees for this class of networks. Our main technical contribution involves extracting, in polynomial time, a constant number of isometric and balanced tree separators from a given CUDG. We do this by extending the celebrated Lipton-Tarjan separator theorem for planar graphs to CUDGs. Our techniques extend to other classes of graphs; for example, for general graphs, we obtain an O(log n)-stretch greedy embedding into O(log2n)-dimensional space. The greedy embeddings constructed by our algorithm can also be viewed as a constant-stretch compact routing scheme in which each node is assigned an O(log3n)-bit label. To the best of our knowledge, this result yields the best known stretch-space trade-off for compact routing on CUDGs. Extensive simulations on random wireless networks indicate that the average routing overhead is about 10%; only few routes have a stretch above 1.5.

In this work we study the problem of determining the throughput capacity of a wireless network. We propose a scheduling algorithm to achieve this capacity within an approximation factor. Our analysis is performed in the physical interference model, where nodes are arbitrarily distributed in Euclidean space. We consider the problem separately from the routing problem and the power control problem, i.e., all requests are single-hop, and all nodes transmit at a fixed power level. The existing solutions to this problem have either concentrated on special-case topologies, or presented optimality guarantees which become arbitrarily bad (linear in the number of nodes) depending on the network's topology. We propose the first scheduling algorithm with approximation guarantee independent of the topology of the network. The algorithm has a constant approximation guarantee for the problem of maximizing the number of links scheduled in one time-slot. Furthermore, we obtain a O(log n) approximation for the problem of minimizing the number of time slots needed to schedule a given set of requests. Simulation results indicate that our algorithm does not only have an exponentially better approximation ratio in theory, but also achieves superior performance in various practical network scenarios. Furthermore, we prove that the analysis of the algorithm is extendable to higher-dimensional Euclidean spaces, and to more realistic bounded-distortion spaces, induced by non-isotropic signal distortions. Finally, we show that it is NP-hard to approximate the scheduling problem to within n1-epsivfactor, for any constant epsiv > 0, in the non-geometric SINR model, in which path-loss is independent of the Euclidean coordinates of the nodes.

Data dissemination in decentralized networks is often realized by using some form of swarming technique. Swarming enables nodes to gather dynamically in order to fulfill a certain task collaboratively and to exchange resources (typically pieces of files or packets of a multimedia data stream). As in most distributed systems, swarming applications face the problem that the nodes in a network have heterogeneous capabilities or act selfishly. We investigate the problem of efficient live data dissemination (e.g., TV streams) in swarms. The live streams should be distributed in such a way that only nodes with sufficiently large contributions to the system are able to fully receive it-even in the presence of freeloading nodes or nodes that upload substantially less than required to sustain the multimedia stream. In contrast, uncooperative nodes cannot properly receive the data stream as they are unable to fill their data buffers in time, incentivizing a fair sharing of resources. If the number of selfish nodes increases, our emulation results reveal that the situation steadily deteriorates for them, while obedient nodes continue to receive virtually all packets in time.

We present a novel clock synchronization algorithm and prove tight upper and lower bounds on the worst-case clock skew that may occur between any two participants in any given distributed system. More importantly, the worst-case clock skew between neighboring nodes is (asymptotically) at most a factor of two larger than the best possible bound. While previous results solely focused on the dependency of the skew bounds on the network diameter, we prove that our techniques are optimal also with respect to the maximum clock drift, the uncertainty in message delays, and the imposed bounds on the clock rates. The presented results all hold in a general model where both the clock drifts and the message delays may vary arbitrarily within pre-specified bounds.

We present a randomized coloring algorithm for the unstructured radio network model, a model comprising autonomous nodes, asynchronous wake-up, no collision detection and an unknown but geometric network topology. The current state-of-the-art coloring algorithm needs with high probability O(Δ ∙ log n) time and uses O(Δ) colors, where n and Δ are the number of nodes in the network and the maximum degree, respectively; this algorithm requires knowledge of a linear bound on n and Δ. We improve this result in three ways: Firstly, we improve the time complexity, instead of the logarithmic factor we just need a polylogarithmic additive term; more specifically, our time complexity is O(Δ + log Δ ∙ log n) given an estimate of n and Δ, and O(Δ + log2 n) without knowledge of Δ. Secondly, our vertex coloring algorithm needs Δ + 1 colors only. Thirdly, our algorithm manages to do a distance-d coloring with asymptotically optimal O(Δ) colors for a constant d.

We present YETI, an Eclipse plug-in providing support for TinyOS development. YETI provides features well-known from development environments for other languages such as syntax highlighting, code completion and error detection. Furthermore, it includes an additional set of tools which are designed to ease the TinyOS development process for both newcomers and experienced developers. The plugin seamlessly integrates with the existing TinyOS toolchain and provides debugging support on the target platform using a JTAG hardware adapter.

Having access to an accurate time is a vital building block in all networks; in wireless sensor networks even more so, because wireless media access or data fusion may depend on it. Starting out with a novel analysis, we show that orthodox clock synchronization algorithms make fundamental mistakes. The state-of-the-art clock synchronization algorithm FTSP exhibits an error that grows exponentially with the size of the network, for instance. Since the involved parameters are small, the error only becomes visible in midsize networks of about 10--20 nodes. In contrast, we present PulseSync, a new clock synchronization algorithm that is asymptotically optimal. We evaluate PulseSync on a Mica2 testbed, and by simulation on larger networks. On a 20 node network, the prototype implementation of PulseSync outperforms FTSP by a factor of 5. Theory and simulation show that for larger networks, PulseSync offers an accuracy which is several orders of magnitude better than FTSP. To round off the presentation, we investigate several optimization issues, e.g. media access and local skew.

In order to be efficient with selfish programmers, a multicore transactional memory (TM) system must be designed such that it is compatible with good programming incentives (GPI), i.e., writing efficient code for the overall system coincides with writing code that optimizes an individual program's performance. By implementing a selfish strategy, we show that under most contention managers (CM) proposed in the literature so far, TM systems are not GPI compatible, whereas a simple randomized CM is GPI compatible.

We present a distributed clock synchronization algorithm that guarantees an exponentially improved bound of O(log D) on the clock skew between neighboring nodes in any graph G of diameter D. In light of the lower bound of Omega(log D/ log log D), this result is almost tight. Moreover, the global clock skew between any two nodes, particularly nodes that are not directly connected, is bounded by O(D), which is optimal up to a constant factor. Our algorithm further ensures that the clock values are always within a linear envelope of real time. A better bound on the accuracy with respect to real time cannot be achieved in the absence of an external timer. These results all hold in a general model where both the clock drifts and the message delays may vary arbitrarily within pre-specified bounds.

We reconsider the problem of geographic routing in wireless ad hoc networks. We are interested in local, memoryless routing algorithms, i.e. each network node bases its routing decision solely on its local view of the network, nodes do not store any message state, and the message itself can only carry information about O(1) nodes. In geographic routing schemes, each network node is assumed to know the coordinates of itself and all adjacent nodes, and each message carries the coordinates of its target. Whereas many of the aspects of geographic routing have already been solved for 2D networks, little is known about higher-dimensional networks. It has been shown only recently that there is in fact no local memoryless routing algorithm for 3D networks that delivers messages deterministically. In this paper, we show that a cubic routing stretch constitutes a lower bound for any local memoryless routing algorithm, and propose and analyze several randomized geographic routing algorithms which work well for 3D network topologies. For unit ball graphs, we present a technique to locally capture the surface of holes in the network, which leads to 3D routing algorithms similar to the greedy-face-greedy approach for 2D networks.

Lucrative incentives in grid computing do not only attract honest participants, but also cheaters. To prevent selfish behavior, verification mechanisms are required. Today's solutions mostly base on redundancy and inherently exhibit a considerable overhead. Often, however, the verification of a result takes much less time than its computation. In this paper we propose a distributed checking scheme that exploits this asymmetry. Our mechanism detects wrong results and excludes cheaters in a distributed manner and hence disburdens the central of the grid server. We show how the verification scheme is used in an application which aims at breaking the discrete logarithm problem by a parallel implementation of the Pollard-p algorithm. Our implementation extends the BOINC server software and is robust to various rational attacks even in the presence of colluders.

Ever larger collections of music are stored on mobile devices. The process of managing these repositories therefore becomes increasingly challenging. In this work we propose to use a map of the "world of music" as a data structure for music exploration and retrieval on mobile devices. We present Mobile Music Explorer---a mobile application, which allows users to create playlists by specifying trajectories on the map and to use similarity based search methods to navigate through their personal music collections. Our navigation methods ensure that any part of the collection can quickly be reached, even for a large set of items. Moreover, we show that the map representation is a natural approach to provide efficient and distributed operation.

This paper studies the fundamental trade-off between communication cost and delay cost arising in various contexts such as control message aggregation or organization theory. An optimization problem is considered where nodes are organized in a tree topology. The nodes seek to minimize the time until the root is informed about their states and to use as few transmissions as possible at the same time. We derive an upper bound on the competitive ratio of O(min(h,c)) where h is the tree's height, and c is the transmission cost per edge. Moreover, we prove that this upper bound is tight in the sense that any oblivious algorithm has a ratio of at least Omega(min(h,c)).

We present a novel distributed algorithm for the maximal independent set (MIS) problem. On growth-bounded graphs (GBG) our deterministic algorithm finishes in O(log* n) time, n being the number of nodes. In light of Linial's Ω(log* n) lower bound our algorithm is asymptotically optimal. Our algorithm answers prominent open problems in the ad hoc/sensor network domain. For instance, it solves the connected dominating set problem for unit disk graphs in O(log* n) time, exponentially faster than the state-of-the-art algorithm. With a new extension our algorithm also computes a delta+1 coloring in O(log* n) time, where delta is the maximum degree of the graph.

This paper studies a virus inoculation game on social networks. A framework is presented which allows the measuring of the windfall of friendship, i.e., how much players benefit if they care about the welfare of their direct neighbors in the social network graph compared to purely selfish environments. We analyze the corresponding equilibria and show that the computation of the worst and best Nash equilibrium is NP-hard. Intriguingly, even though the windfall of friendship can never be negative, the social welfare does not increase monotonically with the extent to which players care for each other. While these phenomena are known on an anecdotal level, our framework allows us to quantify these effects analytically.

Whether local algorithms can compute constant approximations of NP-hard problems is of both practical and theoretical interest. So far, no algorithms achieving this goal are known, as either the approximation ratio or the running time exceed O(1), or the nodes are provided with non-trivial additional information. In this paper, we present the first distributed algorithm approximating a minimum dominating set on a planar graph within a constant factor in constant time. Moreover, the nodes do not need any additional information.

Live streaming is one of many applications where data is continuously created, and has to be quickly distributed among a large number of users. The peer-to-peer paradigm is thereby attracting interest with the prospect of overcoming scalability issues of more centralized approaches. Since data blocks travel along multiple (possibly malicious) peers, authenticating the origin of blocks becomes of prime importance to guarantee safety and reliability. The asymmetry of a single source and an arbitrary number of untrusted receivers requires the use of digital signatures and public key cryptography in general. This paper proposes a new signature scheme for broadcast authentication tailored towards peer-to-peer systems to overcome limitations of traditional approaches based on signature schemes like RSA and DSA, most notably in terms of delays, signature size, and computational complexity. It may further be of practical interest for other real-time applications such as massive multiplayer peer-to-peer gaming.

Ever growing music collections ask for novel ways of organization. The traditional browsing of folder hierarchies or search by title and album tends to be insufficient to maintain an overview of a collection of orders of thousands of tracks. Methods based on song similarity offer an alternative to keyword-based search. In this work we propose to use a high-dimensional map of the "world of music" as a data structure for music retrieval and exploration of personal collections. Our approach does not require expensive analysis of audio signals and scales to hundreds of thousands of tracks. The techniques presented in this work can be used in a variety of applications, ranging from automatic DJs to file sharing on mobile devices. As a concrete example, we have developed a Web-application that allows users to visualize and navigate through their music collections and create playlists by specifying trajectories.

Recently, mobile social software has become an active area of research and development. A multitude of systems have been proposed over the past years that try to follow the success of their Internet bound equivalents. Many mobile solutions try to augment the functionality of existing platforms with location awareness. The price for mobility, however, is typically either the lack of the popular friendship exploration features or the costs involved to access a central server required for this functionality. In this paper, we try to address this issue by introducing a decentralized method that is able to explore the social neighborhood of a user by detecting friends of friends. Rather than only exploiting information about the users of the system, the method relies on real friends, and adequately addresses the arising privacy issues. Moreover, we present VENETA, a mobile social networking platform which, among other features, implements our novel friend of friend detection algorithm.

This paper studies routing schemes and their distributed construction in limited wireless networks, such as sensor or mesh networks. We argue that the connectivity of such networks is well captured by a constant doubling metric and present a constant stretch multicast algorithm through which any network nodeucan send messages to an arbitrary receiver setU. In other words, we describe a distributed approximation algorithm which is only a constant factor off the NP-hard minimum Steiner tree onucupU. As a building block for the multicasting, we construct a 1 + epsiv stretch labeled routing scheme with label size O(log ominus) and storage overhead O(1/epsiv)alpha(log ominus)(O(alpha) + log Delta), where ominus is the diameter of the network, Delta the maximum degree of any network node, andalphaa constant representing the doubling dimension of the network. In addition to unicast and multicast, we present a constant approximation for anycasting on the basis of radic(6)-approximate distance queries. We provide a distributed algorithm to construct the required labeling and routing tables.

In wireless networks mutual interference impairs the quality of received signals and might even prevent the correct reception of messages. It is therefore of paramount importance to dispose of power control and scheduling algorithms, coordinating the transmission of communication requests. We propose a new measure disturbance in order to comprise the intrinsic difficulty of finding a short schedule for a problem instance. Previously known approaches suffer from extremely bad performance in certain network scenarios even if disturbance is low. To overcome this problem, we present a novel scheduling algorithm for which we give analytical worst-case guarantees on its performance. Compared to previously known solutions, the algorithm achieves a speed up, which can be exponential in the size of the network.

Environmental monitoring is one of the driving applications in the domain of sensor networks. The lifetime of such systems is envisioned to exceed several years. To achieve this longevity in unattended operation it is crucial to minimize energy consumption of the battery-powered sensor nodes. This paper proposes Dozer, a data gathering protocol meeting the requirements of periodic data collection and ultra-low power consumption. The protocol comprises MAC-layer, topology control, and routing all coordinated to reduce energy wastage of the communication subsystem. Using a tree-based network structure, packets are reliably routed towards the data sink. Parents thereby schedule precise rendezvous times for all communication with their children. In a deployed network consisting of 40 TinyOS-enabled sensor nodes, Dozer achieves radio duty cycles in the magnitude of 0.2%.

We present BuzzTrack, an email client extension that helps users deal with email overload. This plugin enhances the interface to present messages grouped by topic, instead of the traditional approach of organizing email in folders. We discuss a clustering algorithm that creates the topic-based grouping, and a heuristic for labeling the resulting clusters to summarize their contents. Lastly, we evaluate the clustering scheme in the context of existing work on topic detection and tracking (TDT) for news articles: Our algorithm exhibits similar performance on emails as current work on news text. We believe BuzzTrack's organization structure, which can be obtained at no cost to the end user, will be helpful for managing the massive amounts of email that land in the inbox every day.

In this paper we study the problem of scheduling wireless links in the geometric SINR model, which explicitly uses the fact that nodes are distributed in the Euclidean plane. We present the first NP-completeness proofs in such a model. In particular, we prove two problems to be NP-complete: Scheduling and One-Shot Scheduling. The first problem consists in finding a minimum-length schedule for a given set of links. The second problem receives a weighted set of links as input and consists in finding a maximum-weight subset of links to be scheduled simultaneously in one shot. In addition to the complexity proofs, we devise an approximation algorithm for each problem.

This paper proposes to utilize algorithms from the probabilistic graphical models domain for Peer-to-Peer rating of data items and for computing "social influence" of nodes in a Peer-to-peer social network. We evaluate the practicality of our approach using large- scale simulations over a MSN Live Messenger subgraph consisting of about a million nodes. Our algorithms are general since they can be used for Peer-to-peer monitoring and for the efficient computation of other node ranking methods, such as PageRank and Information Centrality.

We revisit the problem of distributed k-selection where, given a general connected graph of diameter D consisting of n nodes in which each node holds a numeric element, the goal is to determine the kth smallest of these elements. In our model, there is no imposed relation between the magnitude of the stored elements and the number of nodes in the graph. We propose a randomized algorithm whose time complexity is O(DlogD n) with high probability. Additionally, a deterministic algorithm with a worst-case time complexity of O(Dlog2 D n) is presented which considerably improves the best known bound for deterministic algorithms. Moreover, we prove a lower bound of Ω(D logDn) for any randomized or deterministic algorithm, implying that the randomized algorithm is asymptotically optimal.

The virtual world is comprised of data items related to each other in a variety of contexts. Often such relations can be represented as graphs that evolve over time. Examples include social networks, co-authorship graphs, and the world-wide-web. Attempts to model these graphs have introduced the notions of hierarchies and layers, which correspond to taxonomies of the underlying objects, and reasons for object relations, respectively. In this paper we explore these concepts in the process of mining such naturallygrown networks. Based on two sample graphs, we present some evidence that the current models well fit real world networks and provide concrete applications of these findings. In particular, we show how hierarchies can be used for greedy routing and how separation of layers can be used as a preprocessing step to implement a location estimation application.

In this paper, we study distributed approximation algorithms for fault-tolerant clustering in wireless ad hoc and sensor networks. A k-fold dominating set of a graph G = (V,E) is a subset S of V such that every node v \in V \ S has at least k neighbors in S. We study the problem in two network models. In general graphs, for arbitrary parameter t, we propose a distributed algorithm that runs in time O(t^2) and achieves an approximation ratio of O(t\delta^2/t log\delta), where n and \delta denote the number of nodes in the network and the maximal degree, respectively. When the network is modeled as a unit disk graph, we give a probabilistic algorithm that runs in time O(log log n) and achieves an O(1) approximation in expectation. Both algorithms require only small messages of size O(log n) bits.

Abstract — The inherent trade-off between energy-efficiency and rapidity of event dissemination is characteristic for wireless sensor networks. Scarcity of energy renders it necessary for nodes to spend a large portion of their lifetime in an energyefficient sleep mode during which they do neither receive nor send messages. On the other hand, the longer nodes stay in sleep mode, the slower will be the reaction time for disseminating an external event. The trade-off is prominently exhibited during the deployment phase of sensor networks, if some nodes are deployed earlier than others. In this paper, we study this fundamental trade-off by giving a formal model that enables us to compare the performance of different protocols and algorithms. Based on this model, we propose, analyze, and simulate two novel algorithms which significantly outperform existing solutions. I.

We define and study the scheduling complexity in wireless networks, which expresses the theoretically achievable efficiency of MAC layer protocols. Given a set of communication requests in arbitrary networks, the scheduling complexity describes the amount of time required to successfully schedule all requests. The most basic and important network structure in wireless networks being connectivity, we study the scheduling complexity of connectivity, i.e., the minimal amount of time required until a connected structure can be scheduled. In this paper, we prove that the scheduling complexity of connectivity grows only polylogarithmically in the number of nodes. Specifically, we present a novel scheduling algorithm that successfully schedules a strongly connected set of links in time O(log 4 n) even in arbitrary worst-case networks. On the other hand, we prove that standard MAC layer or scheduling protocols can perform much worse. Particularly, any protocol that either employs uniform or linear (a node’s transmit power is proportional to the minimum power required to reach its intended receiver) power assignment has a Ω(n) scheduling complexity in the worst case, even for simple communication requests. In contrast, our polylogarithmic scheduling algorithm allows many concurrent transmission by using an explicitly formulated non-linear power assignment scheme. Our results show that even in large-scale worst-case networks, there is no theoretical scalability problem when it comes to scheduling transmission requests, thus giving an interesting complement to the more pessimistic bounds for the capacity in wireless networks. All results are based on the physical model of communication, which takes into account that the signal-tonoise plus interference ratio (SINR) at a receiver must be above a certain threshold if the transmission is to be received correctly.

Developing algorithms for sensor networks - and proving their correctness and performance - requires simplifying but still realistic models. This paper surveys various models in use today and puts them into perspective. In addition, we propose interesting models which are not widely adopted by the community so far

Until now, the analysis of fault tolerance of peer-to-peer systems usually only covers random faults of some kind. Contrary to traditional algorithmic research, faults as well as joins and leaves occurring in a worst-case manner are hardly considered. In this paper, we devise techniques to build dynamic peer-to-peer systems which remain fully functional in spite of an adversary which continuously adds and removes peers. We exemplify our algorithms on a pancake topology and present a system which maintains peer degree and network diameter O(log n/log log n), where n is the total number of peers in the system

ALGO is a distributed location service to track the position of mobile nodes and to route messages between any two nodes. The lookup of nodes is achieved by searching in a hierarchy of pointers that each node maintains. We show that ALGO has constant stretch for lookup requests. In contrast to previous work, we consider a concurrent setup where nodes are truly mobile and move even while messages are being routed towards them. We prove correctness and efficiency of ALGO and determine the maximum speed at which the nodes might move, which is up to 1/15 of the routing speed. To the best of our knowledge, this is the first work that bounds the node speed, a necessity to prove the success of a lookup algorithm. We verified our theoretical results through extensive simulation and show that the average lookup stretch is around 6.

To date, topology control in wireless ad hoc and sensor networks--the study of how to compute from the given communication network a subgraph with certain beneficial properties .has been considered as a static problem only; the time required to actually schedule the links of a computed topology without message collision was generally ignored. In this paper we analyze topology control in the context of the physical Signal-to-Interference-plus-Noise-Ratio (SINR) model, focusing on the question of how and how fast the links of a resulting topology can actually be realized over time.For this purpose, we define and study a generalized version of the SINR model and obtain theoretical upper bounds on the scheduling complexity of arbitrary topologies in wireless networks. Specifically, we prove that even in worst-case networks, if the signals are transmitted with correctly assigned transmission power levels, the number of time slots required to successfully schedule all links of an arbitrary topology is proportional to the squared logarithm of the number of network nodes times a previously defined static interference measure Interestingly, although originally considered without explicit accounting for signal collision in the SINR model, this static interference measure plays an important role in the analysis of link scheduling with physical link interference. Our result thus bridges the gap between static graph-based interference models and the physical SINR model. Based on these results, we also show that when it comes to scheduling, requiring the communication links to be symmetric may imply significantly higher costs as opposed to topologies allowing unidirectional links.

Peer-to-peer systems (p2p) are highly dynamic in nature. They may consist of millions of peers joining only for a limited period of time, resulting in hundreds of join and leave events per second. In this paper we introduce eQuus, a novel distributed hash table (DHT) suitable for highly dynamic environments. eQuus guarantees that lookups are always fast - in terms of both the delay and the total number of routing hops -, although peers may join and leave the network at any time and concurrently

Coloring the nodes of a graph with a small number of colors is one of the most fundamental problems in theoretical computer science. In this paper, we study graph coloring in a distributed setting. Processors of a distributed system are nodes of an undirected graph G. There is an edge between two nodes whenever the corresponding processors can directly communicate with each other. We assume that distributed coloring algorithms start with an initial m-coloring of G. In the paper, we prove new strong lower bounds for two special kinds of coloring algorithms. For algorithms which run for a single communication round---i.e., every node of the network can only send its initial color to all its neighbors---, we show that the number of colors of the computed coloring has to be at least Ω(Δ2/log2 Δ+ log log m). If such one-round algorithms are iteratively applied to reduce the number of colors step-by-step, we prove a time lower bound of Ω(Δ/log2 Δ+ log*m) to obtain an O(Δ)-coloring. The best previous lower bounds for the two types of algorithms are Ω(log log m) and Ω(log*m), respectively.

Over the last years, game theory has provided great insights into the behavior of distributed systems by modeling the players as utility-maximizing agents. In particular, it has been shown that selfishness causes many systems to perform in a globally suboptimal fashion. Such systems are said to have a large Price of Anarchy. In this paper, we extend this active field of research by allowing some players to be malicious or Byzantine rather than selfish. We ask: What is the impact of Byzantine players on the system's efficiency compared to purely selfish environments or compared to the social optimum? In particular, we introduce the Price of Malice which captures this efficiency degradation. As an example, we analyze the Price of Malice of a game which models the containment of the spread of viruses. In this game, each node can choose whether or not to install anti-virus software. Then, a virus starts from a random node and iteratively infects all neighboring nodes which are not inoculated. We establish various results about this game. For instance, we quantify how much the presence of Byzantine players can deteriorate or---in case of highly risk-averse selfish players---improve the social welfare of the distributed system.

Current peer-to-peer (P2P) systems often suffer from a large fraction of freeriders not contributing any resources to the network. Various mechanisms have been designed to overcome this problem. However, the selfish behavior of peers has aspects which go beyond resource sharing. This paper studies the effects on the topology of a P2P network if peers selfishly select the peers to connect to. In our model, a peer exploits locality properties in order to minimize the latency (or response times) of its lookup operations. At the same time, the peer aims at not having to maintain links to too many other peers in the system. By giving tight bounds on the price of anarchy, we show that the resulting topologies can be much worse than if peers collaborated. Moreover, the network may never stabilize, even in the absence of churn. Finally, we establish the complexity of Nash equilibria in our game theoretic model of P2P networks. Specifically, we prove that it is NP-hard to decide whether our game has a Nash equilibrium and can stabilize.

We present Cryptree, a cryptographic tree structure which facilitates access control in file systems operating on untrusted storage. Cryptree leverages the file system's folder hierarchy to achieve efficient and intuitive, yet simple, access control. The highlights are its ability to recursively grant access to a folder and all its subfolders in constant time, the dynamic inheritance of access rights which inherently prevents scattering of access rights, and the possibility to grant someone access to a file or folder without revealing the identities of other accessors. To reason about and to visualize Cryptree, we introduce the notion of cryptographic links. We describe the Cryptrees we have used to enforce read and write access in our own file system. Finally, we measure the performance of the Cryptree and compare it to other approaches

In this paper we examine the general multi-word lock problem, where processes are allowed to multilock arbitrary registers. Aiming for a highly efficient solution we propose a randomized algorithm which successfully breaks long dependency chains, the crucial factor for slowing down an execution. In the analysis we focus on the 2-word lock problem and show that in this special case an execution of our algorithm takes with high probability at most time O( ∆ 3 log n / log log n), where n is the number of registers and ∆ the maximal number of processes interested in the same register (the contention). Furthermore, we implemented our algorithm for the general multi-word lock problem on an SGI Origin2000 machine, demonstrating that our algorithm is not only of theoretical interest.

Many large-scale networks such as ad hoc and sensor networks, peer-to-peer networks, or the Internet have the property that the number of independent nodes does not grow arbitrarily when looking at neighborhoods of increasing size. Due to this bounded "volume growth," one could expect that distributed algorithms are able to solve many problems more efficiently than on general graphs. The goal of this paper is to help understanding the distributed complexity of problems on "bounded growth" graphs. We show that on the widely used unit disk graph, covering and packing linear programs can be approximated by constant factors in constant time. For a more general network model which is based on the assumption that nodes are in a metric space of constant doubling dimension, we show that in O(log*!n) rounds it is possible to construct a (O(1), O(1))-network decomposition. This results in asymptotically optimal O(log*!n) time algorithms for many important problems.

In this paper, we initiate the study of the approximability of the facility location problem in a distributed setting. In particular, we explore a trade-off between the amount of communication and the resulting approximation ratio. We give a distributed algorithm that, for every constant k, achieves an O(√k(mρ)1/√klog(m+n)) approximation in O(k) communication rounds where message size is bounded to O(log n) bits. The number of facilities and clients are $m$ and n, respectively, and ρ is a coefficient that depends on the cost values of the instance. Our technique is based on a distributed primal-dual approach for approximating a linear program, that does not form a covering or packing program.

We study the distributed complexity of computing a maximal independent set (MIS) in radio networks with completely unknown topology, asynchronous wake-up, and no collision detection mechanism available. Specifically, we propose a novel randomized algorithm that computes a MIS in time O(log2! n) with high probability, where n is the number of nodes in the network. This significantly improving on the best previously known solutions. A lower bound of Ω(log2!n / log log n) given in [11] implies that our algorithm's running time is close to optimal. Our result shows that the harsh radio network model imposes merely an additional O(log n) factor compared to Luby's MIS algorithm in the message passing model. This has important implications in the context of ad hoc and sensor networks whose characteristics are closely captured by the radio network model.

Positioning based on received signal strengths from base stations is highly sensitive to the effects of signal attenuation, reflection, and scattering. Moreover, experimental measurements show that received signal strengths fluctuate significantly over time due to noise and interference. Unlike most of the related work our approach does not rely on the presence of a priori information about base station positions, objects influencing radio signal propagation, and signal propagation characteristics. Instead of trying to compute the current position defined by coordinates, the goal of our system is to infer a user's present logical position. In addition, our system particularly differs from previous work in the chosen statistical approach, which explicitly copes with signal strength fluctuation over time and allows control over system accuracy by means of specification of confidence intervals.

During and immediately after their deployment, ad hoc and sensor networks lack an efficient communication scheme rendering even the most basic network coordination problems difficult. Before any reasonable communication can take place, nodes must come up with an initial structure that can serve as a foundation for more sophisticated algorithms. In this paper, we consider the problem of obtaining a vertex coloring as such an initial structure. We propose an algorithm that works under the unstructured radio network model. This model captures the characteristics of newly deployed ad hoc and sensor networks, i.e. asynchronous wake-up, no collision-detection, and scarce knowledge about the network topology. Our algorithm produces a correct coloring with O(Δ) colors in time O(Δ log n) with high probability in a unit disk graph, where n and Δ are the number of nodes in the network and the maximum degree, respectively. Furthermore, the number of locally used colors depends only on the local node density.

The topology of a wireless multi-hop network can be controlled by varying the transmission power at each node. In this paper, we give a detailed analysis of a cone-based distributed topology-control (CBTC) algorithm. This algorithm does not assume that nodes have GPS information available; rather it depends only on directional information. Roughly speaking, the basic idea of the algorithm is that a node u transmits with the minimum power p/sub u,/spl alpha// required to ensure that in every cone of degree /spl alpha/ around u, there is some node that u can reach with power p/sub u,/spl alpha//. We show that taking /spl alpha/=5/spl pi//6 is a necessary and sufficient condition to guarantee that network connectivity is preserved. More precisely, if there is a path from s to t when every node communicates at maximum power then, if /spl alpha//spl les/5/spl pi//6, there is still a path in the smallest symmetric graph G/sub /spl alpha// containing all edges (u,v) such that u can communicate with v using power p/sub u,/spl alpha//. On the other hand, if /spl alpha/>5/spl pi//6, connectivity is not necessarily preserved. We also propose a set of optimizations that further reduce power consumption and prove that they retain network connectivity. Dynamic reconfiguration in the presence of failures and mobility is also discussed. Simulation results are presented to demonstrate the effectiveness of the algorithm and the optimizations.

Summary form only given. The XTC ad-hoc network topology control algorithm introduced shows three main advantages over previously proposed algorithms. First, it is extremely simple and strictly local. Second, it does not assume the network graph to be a unit disk graph; XTC proves correct also on general weighted network graphs. Third, the algorithm does not require availability of node position information. Instead, XTC operates with a general notion of order over the neighbors' link qualities. In the special case of the network graph being a unit disk graph, the resulting topology proves to have bounded degree, to be a planar graph, and - on average-case graphs - to be a good spanner.

When being deployed, ad-hoc and sensor networks are unstructured and lack an efficient and reliable communication scheme. Hence, the organization of a MAC layer is the primary goal during and immediately after the deployment of such networks. Computing a good initial clustering facilitates this task and is therefore a vital part of the initialization process. A clustering based on a maximal independent set provides several highly desirable properties. Besides yielding a dominating set of good quality, such a clustering avoids interference between clusterheads, thus allowing efficient communication. We propose a novel algorithm that works under a model capturing the characteristics of the initialization phase of unstructured radio networks, i.e., asynchronous wake-up, scarce knowledge about the topology of the network graph, no collision detection, and the hidden terminal problem. We show that even under these hard conditions, the algorithm computes a maximal independent set in polylogarithmic time.

A newly deployed multi-hop radio network is unstructured and lacks a reliable and efficient communication scheme. In this paper, we take a step towards analyzing the problems existing during the initialization phase of ad hoc and sensor networks. Particularly, we model the network as a multi-hop quasi unit disk graph and allow nodes to wake up asynchronously at any time. Further, nodes do not feature a reliable collision detection mechanism, and they have only limited knowledge about the network topology. We show that even for this restricted model, a good clustering can be computed efficiently. Our algorithm efficiently computes an asymptotically optimal clustering. Based on this algorithm, we describe a protocol for quickly establishing synchronized sleep and listen schedule between nodes within a cluster. Additionally, we provide simulation results in a variety of settings.

Topology control in ad-hoc networks tries to lower node energy consumption by reducing transmission power and by confining interference, collisions and consequently retransmissions. Commonly low interference is claimed to be a consequence to sparseness of the resulting topology. In this paper we disprove this implication. In contrast to most of the related work claiming to solve the interference issue by graph sparseness without providing clear argumentation or proofs, we provide a concise and intuitive definition of interference. Based on this definition we show that most currently proposed topology control algorithms do not effectively constrain interference. Furthermore we propose connectivity-preserving an spanner constructions that are interference-minimal.

We introduce the Distributed Approximative System Information Service (DASIS) as a useful scheme to aggregate approximative information on the state of a peer-to-peer system. We present how this service can be integrated into existing peer-to-peer systems, such as Kademlia and Chord. As a sample application, we show how DASIS can be employed for establishing an effective deterministic join algorithm. Through simulation, we demonstrate that the insertion of peers using DASIS information results in a well-balanced system. Moreover, our join algorithm gracefully resolves load imbalances in the system due to unfortunate biased leaves of peers.

We investigate the theoretical limits of positioning algorithms. In particular, we study scenarios where the nodes do not receive anchors directly (multi-hop) and where no physical distance or angle information whatsoever is available (connectivity-based). Since we envision large-scale sensor networks as an application, we are interested in fast, distributed algorithms. As such, we show that plain hop algorithms are not competitive. Instead, for one-dimensional unit disk graphs we present an optimal algorithm HS. For two or more dimensions, we propose an algorithm GHoST which improves upon the basic hop algorithm in theory and in simulations.

We give time lower bounds for the distributed approximation of minimum vertex cover (MVC) and related problems such as minimum dominating set (MDS). In k communication rounds, MVC and MDS can only be approximated by factors Ω(nc/k2/k) and Ω(∆1/k /k) for some constant c, where n and ∆ denote the number of nodes and the largest degree in the graph. The number of rounds required in order to achieve a constant or even only a polylogarithmic approximation ratio is at least Ω ( √ log n / log log n) and Ω(log ∆ / log log ∆). By a simple reduction, the latter lower bounds also hold for the construction of maximal matchings and maximal independent sets.

No abstract available.

Arrow is a prominent distributed protocol which globally orders requests initiated by the nodes in a distributed system. In this paper we present a dynamic analysis of the Arrow protocol. We prove that Arrow is O(log D)-competitive, where D is the diameter of the spanning tree on which Arrow operates. In addition, we show that our analysis is almost tight by proving that for all trees the competitive ratio of Arrow is Ω(log D/log log D).

In this paper we present GOAFR, a new geometric ad-hoc routing algorithm combining greedy and face routing. We evaluate this algorithm by both rigorous analysis and comprehensive simulation. GOAFR is the first ad-hoc algorithm to be both asymptotically optimal and average-case efficient. For our simulations we identify a network density range critical for any routing algorithm. We study a dozen of routing algorithms and show that GOAFR outperforms other prominent algorithms, such as GPSR or AFR.

Abstract.: Finding a small dominating set is one of the most fundamental problems of classical graph theory. In this paper, we present a new fully distributed approximation algorithm based on LP relaxation techniques. For an arbitrary, possibly constant parameter k and maximum node degree $\\Delta$ , our algorithm computes a dominating set of expected size ${\\rm O}(k\\Delta^{2/k}{\\rm log}(\\Delta)\\vert DS_{\\rm {OPT}}\\vert)$ in ${\\rm O}{(k^2)}$ rounds. Each node has to send ${\\rm O}{(k^2\\Delta)}$ messages of size ${\\rm O}({\\rm log}\\Delta)$ . This is the first algorithm which achieves a non-trivial approximation ratio in a constant number of rounds

All too often a seemingly insurmountable divide between theory and practice can be witnessed. In this paper we try to contribute to narrowing this gap in the field of ad-hoc routing. In particular we consider two aspects: We propose a new geometric routing algorithm which is outstandingly efficient on practical average-case networks, however is also in theory asymptotically worst-case optimal. On the other hand we are able to drop the formerly necessary assumption that the distance between network nodes may not fall below a constant value, an assumption that cannot be maintained for practical networks. Abandoning this assumption we identify from a theoretical point of view two fundamentamentally different classes of cost metrics for routing in ad-hoc networks.

This paper examines the role inter-domain topology and routing policy play in the process of delayed Internet routing convergence. In previous work, we showed that the Internet lacks effective inter-domain path fail-over. Unlike circuit-switched networks which exhibit fail-over on the order of milliseconds, we found Internet backbone routers may take tens of minutes to reach a consistent view of the network topology after a fault. In this paper, we expand an our earlier work by exploring the impact of specific Internet provider policies and topologies on the speed of routing convergence. Based on data from the experimental injection and measurement of several hundred thousand inter-domain routing faults, we show that the time for end-to-end Internet convergence depends on the length of the longest possible backup autonomous system path between a source and destination node. We also demonstrate significant variation in the convergence behavior of Internet service providers, with the larger providers exhibiting the fastest convergence latencies. Finally, we discuss possible modifications to BGP and provider routing policies which if deployed, would improve inter-domain routing convergence.

The topology of wireless multihop ad hoc networks can be controlled by varying the transmission power of each node. We propose a simple distributed algorithm where each node makes local decisions about its transmission power and these local decisions collectively guarantee global connectivity. Specifically, based on the directional information, a node grows it transmission power until it finds a neighbor node in every direction. The resulting network topology increases the network lifetime by reducing the transmission power and reduces traffic interference by having low node degrees. Moreover, we show that the routes in the multihop network are efficient in power consumption. We give an approximation scheme in which the power consumption of each route can be made arbitrarily close to the optimal by carefully choosing the parameters. Simulation results demonstrate significant performance improvements.

Distributed queuing is a fundamental problem in distributed computing, arising in a variety of applications. The challenge in designing a distributed queuing algorithm is to minimize message traffic and delay.

The topology of a wireless multi-hop network can be controlled by varying the transmission power at each node. In this paper, we give a detailed analysis of a cone-based distributed topology control algorithm. This algorithm, introduced in [16], does not assume that nodes have GPS information available; rather it depends only on directional information. Roughly speaking, the basic idea of the algorithm is that a node u transmits with the minimum power pu, α required to ensure that in every cone of degree α around u, there is some node that u can reach with power pu, α. We show that taking α = 5π/6 is a necessary and sufficient condition to guarantee that network connectivity is preserved. More precisely, if there is a path from s to t when every node communicates at maximum power then, if α ⪇ 5π/6, there is still a path in the smallest symmetric graph Gα containing all edges (u, v) such that u can communicate with v using power pu, α. On the other hand, if α > 5π/6, connectivity is not necessarily preserved. We also propose a set of optimizations that further reduce power consumption and prove that they retain network connectivity. Dynamic reconfiguration in the presence of failures and mobility is also discussed. Simulation results are presented to demonstrate the effectiveness of the algorithm and the optimizations.