Retrieval-Augmented Generation (RAG) enhances Large Language Models (LLMs) by grounding generation in external knowledge to improve factuality and reduce hallucinations. Yet most deployments assume a centralized corpus, which is infeasible in privacy-aware domains where knowledge remains siloed. This motivates federated RAG (FedRAG), where a central LLM server collaborates with distributed silos without sharing raw documents. In-context RAG violates this requirement by transmitting verbatim documents, whereas parametric RAG encodes documents into light weight adapters that merge with a frozen LLM at inference, avoiding raw-text exchange. We adopt the parametric approach but face two unique challenges induced by FedRAG: high storage and communication from per-document adapters, and destructive aggregation caused by indiscriminately merging multiple adapters. We present FedMosaic, the first federated RAG framework built on parametric adapters. FedMosaic clusters semantically related documents into multi-document adapters with document-specific masks to reduce overhead while preserving specificity, and performs selective adapter aggregation to combine only relevance-aligned, non-conflicting adapters. Experiments show that FedMosaic achieves an average 10.9% higher accuracy than state-of-the-art methods in four categories, while lowering storage costs by 78.8% to 86.3% and communication costs by 91.4%, and never sharing raw documents.

Accurate and efficient multivariate time series (MTS) forecasting is essential for applications such as traffic management and weather prediction, which depend on capturing long-range temporal dependencies and interactions between entities. Existing methods, particularly those based on Transformer architectures, compute pairwise dependencies across all time steps, leading to a computational complexity that scales quadratically with the length of the input. To overcome these challenges, we introduce the Forecaster with Offline Clustering Using Segments (FOCUS), a novel approach to MTS forecasting that simplifies long-range dependency modeling through the use of prototypes extracted via offline clustering. These prototypes encapsulate high-level events in the real-world system underlying the data, summarizing the key characteristics of similar time segments. In the online phase, FOCUS dynamically adapts these patterns to the current input and captures dependencies between the input segment and high-level events, enabling both accurate and efficient forecasting. By identifying prototypes during the offline clustering phase, FOCUS reduces the computational complexity of modeling long-range dependencies in the online phase to linear scaling. Extensive experiments across diverse benchmarks demonstrate that FOCUS achieves state-of-the-art accuracy while significantly reducing computational costs.

Unstructured data, such as images and texts, are increasingly represented as high-dimensional vectors for emerging AI applications like retrieval-augmented generation. A key operation in these applications is querying for vectors that are both semantically similar and temporally relevant. This operation can be formulated as Timestamp Approximate Nearest Neighbor Search (TANNS), where both the vectors and the query incorporate temporal attributes, aiming to retrieve the approximate nearest neighbors valid at the given timestamp. A naive solution is to create separate indexes for each timestamp, which enables accurate and fast searches but incurs high update latency and excessive storage demands. In this paper, we introduce the timestamp graph, a novel structure that supports rapid index updates while minimizing storage costs. Exploiting the temporal locality of changes in valid vectors, our timestamp graph effectively manages a unified index across all historical timestamps, thereby substantially reducing storage overhead. Moreover, we design the historic neighbor tree, which further compresses the space complexity to that of a single-timestamp index. Extensive evaluations on four standard datasets show that our method achieves over 99% accuracy while improving the query efficiency by 4.4× to 138.1× than existing solutions.

Federated learning (FL) has emerged as a prominent distributed learning paradigm to utilize datasets across multiple data providers. In FL, cross-silo data providers often hesitate to share their high-quality dataset unless their data value can be fairly assessed. Shapley value (SV) has been advocated as the standard metric for data valuation in FL due to its desirable properties. However, the computational overhead of SV is prohibitive in practice, as it inherently requires training and evaluating an FL model across an exponential number of dataset combinations. Furthermore, existing solutions fail to achieve high accuracy and efficiency, making practical use of SV still out of reach, because they ignore choosing suitable computation scheme for approximation framework and overlook the property of utility function in FL. We first propose a unified stratified-sampling framework for two widely-used schemes. Then, we analyze and choose the more promising scheme under the FL linear regression assumption. After that, we identify a phenomenon termed key combinations, where only limited dataset combinations have a high-impact on final data value. Building on these insights, we propose a practical approximation algorithm, IPSS, which strategically selects high-impact dataset combinations rather than evaluating all possible combinations, thus substantially reducing time cost with minor approximation error. Furthermore, we conduct extensive evaluations on the FL benchmark datasets to demonstrate that our proposed algorithm outperforms a series of representative baselines in terms of efficiency and effectiveness.

Spatial-temporal data, fundamental to many intelligent applications, reveals dependencies indicating causal links between present measurements at specific locations and historical data at the same or other locations. Within this context, adaptive spatial-temporal graph neural networks (ASTGNNs) have emerged as valuable tools for modelling these dependencies, especially through a data-driven approach rather than pre-defined spatial graphs. While this approach offers higher accuracy, it presents increased computational demands. Addressing this challenge, this paper delves into the concept of localisation within ASTGNNs, introducing an innovative perspective that spatial dependencies should be dynamically evolving over time. We introduce DynAGS, a localised ASTGNN framework aimed at maximising efficiency and accuracy in distributed deployment. This framework integrates dynamic localisation, time-evolving spatial graphs, and personalised localisation, all orchestrated around the Dynamic Graph Generator, a light-weighted central module leveraging cross attention. The central module can integrate historical information in a node-independent manner to enhance the feature representation of nodes at the current moment. This improved feature representation is then used to generate a dynamic sparse graph without the need for costly data exchanges, and it supports personalised localisation. Performance assessments across two core ASTGNN architectures and nine real-world datasets from various applications reveal that DynAGS outshines current benchmarks, underscoring that the dynamic modelling of spatial dependencies can drastically improve model expressibility, flexibility, and system efficiency, especially in distributed settings. © 2025 Owner/Author.

Early-exit networks (EENs), which adapt their computational depths based on input samples, are widely adopted to accelerate inference in edge computing applications. The effectiveness of EENs relies on difficulty-aware training, which tailors shallow exits for simple samples and deep exits for complex ones. However, existing difficulty-aware training schemes assume centralized environments with sufficient data, which become invalid with real-world edge devices. In this paper, we explore difficulty-aware training in a federated manner, where EENs are collaboratively trained on heterogeneous devices. We observe the cross-model exit unalignment phenomenon, a unique problem when aggregating local EENs into a cohesive global model. To address this problem, we design a novel Difficulty-Aligned Reverse Knowledge Distillation scheme named DarkDistill that preserves the difficulty-specific specialization for aggregating heterogeneous local models. Instead of direct parameter averaging, it trains difficulty-conditional data generators, and selectively transfers generated knowledge of specific difficulty among matched exits of heterogeneous EENs. Evaluations show that DarkDistill outperforms the state-of-the-arts in both full-parameter and parameter-efficient fine-tuning of EENs.

Clustered Federated Learning (CFL) is an emerging paradigm to extract insights from data on IoT devices. Through iterative client clustering and model aggregation, CFL adeptly manages data heterogeneity, ensures privacy, and delivers personalized models to heterogeneous devices. Traditional CFL approaches, which operate synchronously, suffer from prolonged latency for waiting slow devices during clustering and aggregation. This paper advocates a shift to asynchronous CFL, allowing the server to process client updates as they arrive. This shift enhances training efficiency yet introduces complexities to the iterative training cycle. To this end, we present CASA, a novel CFL scheme for Clustering-Aggregation Synergy under Asynchrony. Built upon a holistic theoretical understanding of asynchrony's impact on CFL, CASA adopts a bi-level asynchronous aggregation method and a buffer-aided dynamic clustering strategy to harmonize between clustering and aggregation. Extensive evaluations on standard benchmarks show that CASA outperforms representative baselines in model accuracy and achieves 2.28-6.49× higher convergence speed.

Graph-based methods have witnessed tremendous success in traffic prediction, largely attributed to their superior ability in capturing and modeling spatial dependencies. However, urban-scale traffic data are usually distributed among various owners, limited in sharing due to privacy restrictions. This fragmentation of data severely hinders interaction across clients, impeding the utilization of inter-client spatial dependencies. Existing studies have yet to address this non-trivial issue, thereby leading to sub-optimal performance. To fill this gap, we propose FedGTP, a new federated graph-based traffic prediction framework that promotes adaptive exploitation of inter-client spatial dependencies to recover close-to-optimal performance complying with privacy regulations like GDPR. We validate FedGTP via large-scale application-driven experiments on real-world datasets. Extensive baseline comparison, ablation study and case study demonstrate that FedGTP indeed surpasses existing methods through fully recovering inter-client spatial dependencies, achieving 21.08%, 13.48%, 19.90% decrease on RMSE, MAE and MAPE, respectively. Our code is available at https://github.com/LarryHawkingYoung/KDD2024_FedGTP

Cross-domain recommendation (CDR) is a widely used methodology in recommender systems to combat data sparsity.It leverages user data across different domains or platforms for providing personalized recommendations.Traditional CDR assumes user preferences and behavior data can be shared freely among cloud and users, which is now impractical due to strict restrictions of data privacy.In this paper, we propose a Deployment-friendly Cloud-Device Collaboration framework for Cross-Domain Recommendation (De-CoCDR).It splits CDR into a two-stage recommendation model through cloud-device collaborations, i.e., item-recall on cloud and item re-ranking on device.This design enables effective CDR while preserving data privacy for both the cloud and the device.Extensive offline and online experiments are conducted to validate the effectiveness of DeCoCDR.In offline experiments, DeCoCDR outperformed the state-of-the-arts in three large datasets.While in real-world deployment, DeCoCDR improved the conversion rate by 45.3% compared with the baseline.

Shared mobility leverages under-utilized vehicles to offer on-demand transport services by sharing vehicles among users. It strives to match supply with demand via a series of data-intensive operations such as supply prediction and task assignment. However, its full potential is often compromised in practice as most shared mobility platforms operate in isolation, leading to sub-optimal resource utilization. In this demonstration, we advocate a federated approach to shared mobility, which enhances its effectiveness by enabling optimizations across platforms while retaining their autonomy. We develop privacy-preserving operators and incentive mechanisms dedicated to supply prediction and task assignment in shared mobility and implement generic interfaces that support diverse prediction and assignment algorithms. We showcase the shared mobility system with real-world ride-hailing applications.

Data federation has emerged as a novel database system enabling collaborative queries across mutually distrusted data owners. Federated equi-join, a commonly used operation in data federation, combines relations from distinct data owners while preserving their data privacy. Due to the wide applications of this query, many solutions to federated equi-joins have been proposed. However, it is still challenging for practitioners to choose the most appropriate algorithm due to various reasons, including incomplete evaluation protocols (e.g., lack of evaluating multi-way equi-joins), under-explored performance metric (main memory usage), and absence of a standardized comparison. Motivated by this reason, this paper conducts a comprehensive experimental study and builds a new benchmark, called${\sf FEJ-Bench}$, for federated equi-joins. The experimental study and the benchmark consist of eight state-of-the-art algorithms and five datasets. Our evaluation reveals the query efficiency ranking, its impact factors, and potential research opportunities. Finally, we open-source${\sf FEJ-Bench}$on GitHub, which is the first benchmark for federated equi-joins. Our findings aim to guide researchers and practitioners in deploying federated equi-joins in practice.

Federated Trajectory Matching (FTM) is gaining increasing importance in big trajectory data analytics, supporting diverse applications such as public health, law enforcement, and emergency response. FTM retrieves trajectories that match with a query trajectory from a large-scale trajectory database, while safeguarding the privacy of trajectories in both the query and the database. A naive solution to FTM is to process the query through Secure Multi-party Computation (SMC) across the entire database, which is inherently secure yet inevitably slow due to the massive secure operations. A promising acceleration strategy is to filter irrelevant trajectories from the database based on the query, thus reducing the SMC operations. However, a key challenge is how to publish the query in a way that both preserves privacy and enables efficient trajectory filtering. In this paper, we design${\sf GIST}$, a novel framework for efficient Federated Trajectory Matching.${\sf GIST}$is grounded in Geo-Indistinguishability, a privacy criterion dedicated to locations. It employs a new privacy mechanism for the query that facilitates efficient trajectory filtering. We theoretically prove the privacy guarantee of the mechanism and the accuracy of the filtering strategy of${\sf GIST}$. Extensive evaluations on five real datasets show that${\sf GIST}$is significantly faster and incurs up to 2 orders of magnitude lower communication cost than the state-of-the-arts.

Multi-robot systems are deployed in modern warehouses to reduce operational cost. The robots are tasked to deliver items stored on racks to pickers for fast distribution. A central algorithmic problem is collision-aware route planning, which aims to plan shortest routes for robots to deliver racks while avoiding collision with racks, pickers, and other robots. Prior solutions are inefficient in real-world warehouses, where route planning requests emerge online and at large scale. In this paper, we identify collision judgement in grid-based warehouse representation as the primary efficiency bottleneck, and propose a novel Strip-based Route Planning framework (SRP). Specifically, we exploit the regularity in warehouse layouts, and aggregate grids into strips. The strip-based representation also converts collisions of 3-dimensional (2-dimensional space and 1-dimensional time) routes into 2-dimensional (1-dimensional space and 1-dimensional time) segment intersections, which can be fast checked via computational geometry. We further accelerate the collision judgement via indexing on segments within strips. Theoretical analysis shows a reduction of time complexity from square to linear-logarithmic. Experimental results on datasets collected from real-world robotized warehouses show that our SRP is up to 227× faster than existing methods.

Federated learning (FL) has emerged as a popular machine learning paradigm recently. Compared with traditional distributed learning, its unique challenges mainly lie in communication efficiency and non-IID (heterogeneous data) problem. While the widely adopted framework FedAvg can reduce communication overhead significantly, its effectiveness on non-IID data still lacks exploration. In this paper, we study the non-IID problem of FL from the perspective of domain adaptation. We propose a distribution regularization for FL on non-IID data such that the discrepancy of data distributions between clients is reduced. To further reduce the communication cost, we devise two novel distributed learning algorithms, namely rFedAvg and rFedAvg+, for efficiently learning with the distribution regularization. More importantly, we theoretically establish their convergence for strongly convex objectives. Extensive experiments on 4 datasets with both CNN and LSTM as learning models verify the effectiveness and efficiency of the proposed algorithms.

Online real estate platforms are gaining increasing popularity, where a central issue is to match brokers with clients for potential housing transactions. Mainstream platforms match brokers via top-k recommendation. Yet we observe through extensive data analysis that such top-k recommendation tends to overload the top brokers, which notably degrades their service quality. In this paper, we propose to avoid such overloading in broker matching via the paradigm shift from recommendation to assignment. To this end, we design learned assignment with contextual bandits (LACB), a data-driven capacity-aware assignment scheme for broker matching which estimates broker-specific workload capacity in an online fashion and assigns brokers to clients from a global perspective to maximize the overall service quality. Extensive evaluations on synthetic and real-world datasets from an industrial online real estate platform validate the efficiency and effectiveness of our solution.

Spatial-temporal graph models are prevailing for abstracting and modelling spatial and temporal dependencies. In this work, we ask the following question: whether and to what extent can we localise spatial-temporal graph models? We limit our scope to adaptive spatial-temporal graph neural networks (ASTGNNs), the state-of-the-art model architecture. Our approach to localisation involves sparsifying the spatial graph adjacency matrices. To this end, we propose Adaptive Graph Sparsification (AGS), a graph sparsification algorithm which successfully enables the localisation of ASTGNNs to an extreme extent (fully localisation). We apply AGS to two distinct ASTGNN architectures and nine spatial-temporal datasets. Intriguingly, we observe that spatial graphs in ASTGNNs can be sparsified by over 99.5% without any decline in test accuracy. Furthermore, even when ASTGNNs are fully localised, becoming graph-less and purely temporal, we record no drop in accuracy for the majority of tested datasets, with only minor accuracy deterioration observed in the remaining datasets. However, when the partially or fully localised ASTGNNs are reinitialised and retrained on the same data, there is a considerable and consistent drop in accuracy. Based on these observations, we reckon that (i) in the tested data, the information provided by the spatial dependencies is primarily included in the information provided by the temporal dependencies and, thus, can be essentially ignored for inference; and (ii) although the spatial dependencies provide redundant information, it is vital for the effective training of ASTGNNs and thus cannot be ignored during training. Furthermore, the localisation of ASTGNNs holds the potential to reduce the heavy computation overhead required on large-scale spatial-temporal data and further enable the distributed deployment of ASTGNNs.

Personalized federated learning collaboratively trains client-specific models, which holds potential for various mobile and IoT applications with heterogeneous data. However, existing solutions are vulnerable to distribution shifts between training and test data, and involve high training workloads on local devices. These two shortcomings hinder the practical usage of personalized federated learning on real-world mobile applications. To overcome these drawbacks, we explore efficient shift-robust personalization for federated learning. The principle is to hitchhike the global model to improve the shift-robustness of personalized models with minimal extra training overhead. To this end, we present DM-PFL, a novel framework that utilizes a dual masking mechanism to train both global and personalized models with weight-level parameter sharing and end-to-end sparse training. Evaluations on various datasets show that our methods not only improve the test accuracy in presence of test-time distribution shifts but also save the communication and computation costs compared to state-of-the-art personalized federated learning schemes.

Range aggregation is a primitive operation in spatial data applications and there is a growing demand to support such operations over a data federation, where the entire spatial data are separately held by multiple data providers (a.k.a., data silos). Data federations notably increase the amount of data available for data-intensive applications such as smart mobility planning and public health emergency responses. Yet they also challenge the conventional implementation of range aggregation queries because the raw data cannot be shared within the federation and the data partition at each data silo is fixed during query processing. These constraints limit the design space of distributed range aggregation query processing. In this work, we propose approximate algorithms for efficient range aggregation over spatial data federation. We devise novel single-silo sampling algorithms that process queries in parallel and design a level sampling based algorithm which reduces the time complexity of local queries at each data silo to O(log 1/), where is the approximation ratio of the accuracy guarantee. Extensive evaluations with real-world data show that compared with state-of-the-arts, our solutions reduce the time cost and communication cost by up to 85.1x and 5.5x respectively, with average approximate errors of below 2.8%.

Skyline is a primitive operation in multi-objective decision applications and there is a growing demand to support such operations over a data federation, where the entire dataset is separately held by multiple data providers (a.k.a., silos). Data federations notably increase the amount of data available for data-intensive applications such as commercial recommendation and location based services. Yet they also challenge the conventional implementation of skyline queries because the raw data cannot be shared within the federation and the secure computation cross silos can be two or three orders of magnitude slower than plaintext computation. These constraints render existing solutions inefficient on data federation. In this work, we propose a novel local dominance based framework for efficient skyline queries over a vertical data federation. We decompose the skyline query into plaintext local dominance computations and secure result aggregations, which can perform as many computations in plaintext as possible without compromising security. We further propose a dedicate private set intersection based algorithm to accelerate the query processing. Extensive evaluations on both synthetic and real-world datasets show that compared with general-purpose secure multi-party computation techniques, our solutions reduce the time cost by up to 35.4× and communication cost by two orders of magnitude respectively.

The insurance industry is shifting their sales mode from offline to online, in expectation to reach massive potential customers in the digitization era. Due to the complexity and the nature of insurance products, a cost-effective online sales solution is to exploit chatbot AI to raise customers’ attention and pass those with interests to human agents for further sales. For high response and conversion rates of customers, it is crucial for the chatbot to initiate a conversation with personalized opening sentences, which are generated with user-specific topic selection and ordering. Such personalized opening sentence generation is challenging because (i) there are limited historical samples for conversation topic recommendation in online insurance sales and (ii) existing text generation schemes often fail to support customized topic ordering based on user preferences. We design POSGen, a personalized opening sentence generation scheme dedicated for online insurance sales. It transfers user embeddings learned from auxiliary online user behaviours to enhance conversation topic recommendation, and exploits a context management unit to arrange the recommended topics in user-specific ordering for opening sentence generation. POSGen is deployed on a real-world online insurance platform. It achieves 2.33x total insurance premium improvement through a two-month global test.

Robotized warehouses are deployed to automatically distribute millions of items brought by the massive logistic orders from e-commerce. A key to automated item distribution is to plan paths for robots, also known as task planning, where each task is to deliver racks with items to pickers for processing and then return the rack back. Prior solutions are unfit for large-scale robotized warehouses due to the inflexibility to time-varying item arrivals and the low efficiency for high throughput. In this paper, we propose a new task planning problem called TPRW, which aims to minimize the end-to-end makespan that incorporates the entire item distribution pipeline, known as a fulfilment cycle. Direct extensions from state-of-the-art path finding methods are ineffective to solve the TPRW problem because they fail to adapt to the bottleneck variations of fulfillment cycles. In response, we propose Efficient Adaptive Task Planning, a framework for large-scale robotized warehouses with time-varying item arrivals. It adaptively selects racks to fulfill at each timestamp via rein-forcement learning, accounting for the time-varying bottleneck of the fulfillment cycles. Then it finds paths for robots to transport the selected racks. The framework adopts a series of efficient optimizations on both time and memory to handle large-scale item throughput. Evaluations on both synthesized and real data show an improvement of 37.1% in effectiveness and 75.5% in efficiency over the state-of-the-arts.

Data federations notably increase the amount of data available for data-intensive applications such as smart mobility planning and public health emergency responses. Yet they also challenge the conventional implementation of range aggregation queries because the raw data cannot be shared within the federation and the data partition at each data silo is fixed during query processing. In this work, we propose the first-of-its-kind approximate algorithms for efficient range aggregation over spatial data federation. We devise novel single-silo sampling algorithms that process queries in parallel and design a level sampling based algorithm which reduces the time complexity of local queries at each data silo to$O(\log\frac{1}{\epsilon})$, where ∊ is the approximation ratio of the accuracy guarantee. Extensive experiments on real-world dataset validate the efficiency and effectiveness of the solutions.

The lottery ticket hypothesis (LTH) states that a randomly initialized dense network contains sub-networks that can be trained in isolation to the performance of the dense network. In this paper, to achieve rapid learning with less computational cost, we explore LTH in the context of meta learning. First, we experimentally show that there are sparse sub-networks, known as meta winning tickets, which can be meta-trained to few-shot classification accuracy to the original backbone. The application of LTH in meta learning enables the adaptation of meta-trained networks on various IoT devices with fewer computation. However, the status quo to identify winning tickets requires iterative training and pruning, which is particularly expensive for finding meta winning tickets. To this end, then we investigate the inter- and intra-layer patterns among different meta winning tickets, and propose a scheme for early detection of a meta winning ticket. The proposed scheme enables efficient training in resource-limited devices. Besides, it also designs a lightweight solution to search the meta winning ticket. Evaluations on standard few-shot classification benchmarks show that we can find competitive meta winning tickets with 20% weights of the original backbone, while incurring only 8%-14% (Conv-4) and 19%-29% (ResNet-12) computation overhead (measured by FLOPs) of the standard winning ticket finding scheme.

Data collected by IoT devices are often private and have a large diversity across users. Therefore, learning requires pre-training a model with available representative data samples, deploying the pre-trained model on IoT devices, and adapting the deployed model on the device with local data. Such an on-device adaption for deep learning empowered applications demands data and memory efficiency. However, existing gradient-based meta learning schemes fail to support memory-efficient adaptation. To this end, we propose p-Meta, a new meta learning method that enforces structure-wise partial parameter updates while ensuring fast generalization to unseen tasks. Evaluations on few-shot image classification and reinforcement learning tasks show that p-Meta not only improves the accuracy but also substantially reduces the peak dynamic memory by a factor of 2.5 on average compared to state-of-the-art few-shot adaptation methods.

Learning based order dispatching has witnessed tremendous success in ride hailing. However, the success halts within individual ride hailing platforms because sharing raw order dispatching data across platforms may leak user privacy and business secrets. Such data isolation not only impairs user experience but also decreases the potential revenues of the platforms. In this paper, we advocate federated order dispatching for cross-platform ride hailing, where multiple platforms collaboratively make dispatching decisions without sharing their local data. Realizing this concept calls for new federated learning strategies that tackle the unique challenges on effectiveness, privacy and efficiency in the context of order dispatching. In response, we devise Federated Learning-to-Dispatch (Fed-LTD), a framework that allows effective order dispatching by sharing both dispatching models and decisions while providing privacy protection of raw data and high efficiency. We validate Fed-LTD via large-scale trace-driven experiments with Didi GAIA dataset. Extensive evaluations show that Fed-LTD outperforms single-platform order dispatching by 10.24% to 54.07% in terms of total revenue.

The increasing concerns on data security limit the sharing of data distributedly stored at multiple data owners and impede the scale of spatial queries over big urban data. In response, data federation systems have emerged to perform secure queries across multiple data owners leveraging secure multi-party computation. However, existing systems are designed for relational data. They are highly inefficient on spatial queries and limited in usability. In this demonstration, we introduce Hu-Fu, the first data federation system for secure spatial queries with high efficiency and usability. Hu-Fu is designed from the perspectives of the query user and the data owner for high usability and decomposes a spatial query into as many plaintext operators and as few secure operators as possible for high efficiency. We demonstrate the deployment and usage of Hu-Fu via cross-company taxi-calling, a popular smart city application.

Data isolation has become an obstacle to scale up query processing over big data, since sharing raw data among data owners is often prohibitive due to security concerns. A promising solution is to perform secure queries over a federation of multiple data owners leveraging secure multi-party computation (SMC) techniques, as evidenced by recent federation work over relational data. However, existing solutions are highly inefficient on spatial queries due to excessive secure distance operations for query processing and their usage of general-purpose SMC libraries for secure operation implementation. In this paper, we propose Hu-Fu, the first system for efficient and secure spatial query processing on a data federation. The idea is to decompose the secure processing of a spatial query into as many plaintext operations and as few secure operations as possible, where fewer secure operators are involved and all secure operators are implemented dedicatedly. As a working system, Hu-Fu supports not only query input in native SQL, but also heterogeneous spatial databases ( e.g. , PostGIS, Simba, GeoMesa, and SpatialHadoop) at the backend. Extensive experiments show that Hu-Fu usually outperforms the state-of-the-arts in running time and communication cost while guaranteeing security.

There has been a dramatic growth of shared mobility applications such as ride-sharing, food delivery, and crowdsourced parcel delivery. Shared mobility refers to transportation services that are shared among users, where a central issue is route planning . Given a set of workers and requests, route planning finds for each worker a route, i.e., a sequence of locations to pick up and drop off passengers/parcels that arrive from time to time, with different optimization objectives. Previous studies lack practicability due to their conflicted objectives and inefficiency in inserting a new request into a route, a basic operation called insertion . In addition, previous route planning solutions fail to exploit the appearance patterns of future requests hidden in historical data for optimization. In this paper, we present a unified formulation of route planning called URPSM. It has a well-defined parameterized objective function which eliminates the contradicted objectives in previous studies and enables flexible multi-objective route planning for shared mobility. We propose two insertion-based frameworks to solve the URPSM problem. The first is built upon the plain-insertion widely used in prior studies, which processes online requests only, whereas the second relies on a new insertion operator called prophet-insertion that handles both online and predicted requests. Novel dynamic programming algorithms are designed to accelerate both insertions to only linear time. Theoretical analysis shows that no online algorithm can have a constant competitive ratio for the URPSM problem under the competitive analysis model, yet our prophet-insertion-based framework can achieve a constant optimality ratio under the instance-optimality model. Extensive experimental results on real datasets show that our insertion-based solutions outperform the state-of-the-art algorithms in both effectiveness and efficiency by a large margin (e.g., up to 30 \( \times \) more effective in the objective and up to 20 \( \times \) faster).

Adapting neural networks to unseen tasks with few training samples on resource-constrained devices benefits various Internet-of-Things applications. Such neural networks should learn the new tasks in few shots and be compact in size. Meta-learning enables few-shot learning, yet the meta-trained networks can be over-parameterised. However, naive combination of standard compression techniques like network pruning with meta-learning jeopardises the ability for fast adaptation. In this work, we propose adaptation-aware network pruning (ANP), a novel pruning scheme that works with existing meta-learning methods for a compact network capable of fast adaptation. ANP uses weight importance metric that is based on the sensitivity of the meta-objective rather than the conventional loss function, and adopts approximation of derivatives and layer-wise pruning techniques to reduce the overhead of computing the new importance metric. Evaluations on few-shot classification benchmarks show that ANP can prune meta-trained convolutional and residual networks by 85% without affecting their fast adaptation.

Many mobile applications demand selective execution of multiple correlated deep learning inference tasks on resource-constrained platforms. Given a set of deep neural networks, each pre-trained for a single task, it is desired that executing arbitrary combinations of tasks yields minimal computation cost. Pruning each network separately yields suboptimal computation cost due to task relatedness. A promising remedy is to merge the networks into a multitask network to eliminate redundancy across tasks before network pruning. However, pruning a multitask network combined by existing network merging schemes cannot minimise the computation cost of every task combination because they do not consider such a future pruning. To this end, we theoretically identify the conditions such that pruning a multitask network minimises the computation of all task combinations. On this basis, we propose Pruning-Aware Merging (PAM), a heuristic network merging scheme to construct a multitask network that approximates these conditions. The merged network is then ready to be further pruned by existing network pruning methods. Evaluations with different pruning schemes, datasets, and network architectures show that PAM achieves up to 4.87x less computation against the baseline without network merging, and up to 2.01x less computation against the baseline with a state-of-the-art network merging scheme.

Ride hailing is a widespread shared mobility application where the central issue is to assign taxi requests to drivers with various objectives. Despite extensive research on task assignment in ride hailing, the fairness of earnings among drivers is largely neglected. Pioneer studies on fair task assignment in ride hailing are ineffective and inefficient due to their myopic optimization perspective and time-consuming assignment techniques. In this work, we propose LAF, an effective and efficient task assignment scheme that optimizes both utility and fairness. We adopt reinforcement learning to make assignments in a holistic manner and propose a set of acceleration techniques to enable fast fair assignment on large-scale data. Experiments show that LAF outperforms the state-of-the-arts by up to 86.7%, 29.1%, 797% on fairness, utility and efficiency, respectively.

Sustained work enthusiasms of drivers are crucial for the success of large-scale ride-hailing platforms. In this paper, we conduct the first-of-its-kind exploration to encourage active participation of drivers via competition. We design Arena, a competition where drivers compete for prizes via completing more trips. Through a pilot study covering over 2,600 participants, we uncover the easy-win problem, an overlooked and serious issue in competition design for real-world drivers. It refers to situations where one competitor does not show up during competition whereas the other easily wins. To solve the easy-win problem without impairing motivation of drivers, we devise a novel prediction-based matchmaking framework. On observing that no-shows are highly correlated to the online time of drivers during competition, we propose to identify potential no-shows by predicting drivers' online time and avoid matching potential noshow drivers with drivers that will show up so as to reduce easy-wins. We conduct large-scale experiments based on real competition data involving over 10,000 drivers. The results show that our prediction-based matchmaking scheme can effectively reduce the ratio of easy-wins.

Spatial crowdsourcing has stimulated various new applications such as taxi calling and food delivery. A key enabler for these spatial crowdsourcing based applications is to plan routes for crowd workers to execute tasks given diverse requirements of workers and the spatial crowdsourcing platform. Despite extensive studies on task planning in spatial crowdsourcing, few have accounted for the location privacy of tasks, which may be misused by an untrustworthy platform. In this paper, we explore efficient task planning for workers while protecting the locations of tasks. Specifically, we define the Privacy-Preserving Task Planning (PPTP) problem, which aims at both total revenue maximization of the platform and differential privacy of task locations. We first apply the Laplacian mechanism to protect location privacy, and analyze its impact on the total revenue. Then we propose an effective and efficient task planning algorithm for the PPTP problem. Extensive experiments on both synthetic and real datasets validate the advantages of our algorithm in terms of total revenue and time cost.

With spatial crowdsourcing applications such as Uber and Waze deeply penetrated into everyday life, there is a growing concern to protect user privacy in spatial crowdsourcing. Particularly, locations of workers and tasks should be properly processed via certain privacy mechanism before reporting to the untrusted spatial crowdsourcing server for task assignment. Privacy mechanisms typically permute the location information, which tends to make task assignment ineffective. Prior studies only provide guarantees on privacy protection without assuring the effectiveness of task assignment. In this paper, we investigate privacy protection for online task assignment with the objective of minimizing the total distance, an important task assignment formulation in spatial crowdsourcing. We design a novel privacy mechanism based on Hierarchically Well-Separated Trees (HSTs). We prove that the mechanism is ε-Geo-Indistinguishable and show that there is a task assignment algorithm with a competitive ratio of O(ε1/4log N log2k), where ε is the privacy budget, N is the number of predefined points on the HST, and k is the matching size. Extensive experiments on synthetic and real datasets show that online task assignment under our privacy mechanism is notably more effective in terms of total distance than under prior differentially private mechanisms.

There is a growing trend to deploy deep neural networks at the edge for high-accuracy, real-time data mining and user interaction. Applications such as speech recognition and language understanding often apply a deep neural network to encode an input sequence and then use a decoder to generate the output sequence. A promising technique to accelerate these applications on resource-constrained devices is network pruning, which compresses the size of the deep neural network without severe drop in inference accuracy. However, we observe that although existing network pruning algorithms prove effective to speed up the prior deep neural network, they lead to dramatic slowdown of the subsequent decoding and may not always reduce the overall latency of the entire application. To rectify such drawbacks, we propose entropy-based pruning, a new regularizer that can be seamlessly integrated into existing network pruning algorithms. Our key theoretical insight is that reducing the information entropy of the deep neural network outputs decreases the upper bound of the subsequent decoding search space. We validate our solution with two state-of-the-art network pruning algorithms on two model architectures. Experimental results show that compared with existing network pruning algorithms, our entropy-based pruning method notably suppresses and even eliminates the increase of decoding time, and achieves shorter overall latency with only negligible extra accuracy loss in the applications.

On-demand taxi-calling platforms often ignore the social engagement of individual drivers. The lack of social incentives impairs the work enthusiasms of drivers and will affect the quality of service. In this paper, we propose to form teams among drivers to promote participation. A team consists of a leader and multiple members, which acts as the basis for various group-based incentives such as competition. We define the Recommendation-based Team Formation (RTF) problem to form as many teams as possible while accounting for the choices of drivers. The RTF problem is challenging. It needs both accurate recommendation and coordination among recommendations, since each driver can be in at most one team. To solve the RTF problem, we devise a Recommendation-Matrix-Based Framework (RMBF). It first estimates the acceptance probability of recommendations and then derives a recommendation matrix to maximize the number of formed teams from a global view. We conduct trace-driven simulations using real data covering over 64,000 drivers and deploy our solution on a large on-demand taxi-calling platform for online evaluations. Experimental results show that RMBF outperforms the greedy-based strategy by forming up to 20% and 12.4% teams in trace-driven simulations and online evaluations, and the drivers who form teams and are involved in the competition have more service time, number of finished orders and income.

The last decade has witnessed the emergence and popularity of event-based social networks (EBSNs), which extend online social networks to the physical world. Fundamental on EBSN platforms is to appropriately assign EBSN users to events they are interested to attend, known as event-participant arrangement. Previous event-participant arrangement studies either fail to avoid conflicts among events or ignore the social interactions among participants. In this work, we propose a new event-participant arrangement problem called Interaction-aware Global Event-Participant Arrangement (IGEPA). It globally optimizes arrangements between events and participants to avoid conflicts in events, and not only accounts for user interests, but also encourages socially active participants to join. To solve the IGEPA problem, we design an approximation algorithm which has an approximation ratio of at least 1\4. Experimental results validate the effectiveness of our solution.

A crowdsourcing task in real-world applications often consists of thousands of atomic tasks. A common practice to distribute a large-scale crowdsourcing task is to pack atomic tasks into task bins and send to crowd workers in batches. It is challenging to decompose a large-scale crowdsourcing task into task bins to ensure reliability at a minimal total cost. In this paper, we propose the Smart Large-scAle task DEcomposer (SLADE) problem, which aims to decompose a large-scale crowdsourcing task to achieve the desired reliability at a minimal cost. We prove its NP-hardness and study two variants of the problem. For the homogeneous SLADE problem, we propose a greedy algorithm and an approximation framework using an optimal priority queue (OPQ) structure with provable approximation ratio. For the heterogeneous SLADE problem, we extend this framework and prove its approximation guarantee. Extensive experiments validate the effectiveness and efficiency of the solutions.

Spatial crowdsourcing brings in a new approach for social media and location-based services (LBS) to collect location specific information via mobile users. For example, when a user checks in at a shop on Facebook, he will immediately receive and is asked to complete a set of tasks such as "what is the opening hour of the shop". It is non-trivial to complete a set of tasks timely and accurately via spatial crowdsourcing. Since workers in spatial crowdsourcing are often transient and limited in number, these social media platforms need to properly allocate workers within the set of tasks such that all tasks are completed (i) with high quality and (ii) with a minimal latency (estimated by the arriving index of the last recruited worker). Solutions to quality and latency control in traditional crowdsourcing are inapplicable in this problem because they either assume sufficient workers or ignore the spatiotemporal factors. In this work, we define the Latency-oriented Task Completion (LTC) problem, which trades off quality and latency (number of workers) of task completion in spatial crowdsourcing. We prove that the LTC problem is NP-hard. We first devise a minimum-cost-flow based algorithm with a constant approximation ratio for the LTC problem in the offline scenario, where all information is known a prior. Then we study the more practical online scenario of the LTC problem, where workers appear dynamically and the platform needs to arrange tasks for each worker immediately based on partial information. We design two greedy-based algorithms with competitive ratio guarantees to solve the LTC problem in the online scenario. Finally, we validate the effectiveness and efficiency of the proposed solutions through extensive evaluations on both synthetic and real-world datasets.

Cars of the future have been predicted as shared and electric. There has been a rapid growth in electric vehicle (EV) sharing services worldwide in recent years. For EV-sharing platforms to excel, it is essential for them to offer private charging infrastructure for exclusive use that meets the charging demand of their clients. Particularly, they need to plan not only the places to build charging stations, but also the amounts of chargers per station, to maximally satisfy the requirements on global charging coverage and local charging demand. Existing research efforts are either inapplicable for their different problem formulations or are at a coarse granularity. In this paper, we formulate the \underlineE lectric \underlineV ehicle \underlineC harger \underlineP lanning (EVCP) problem especially for EV-sharing. We prove that the \shortpro problem is NP-hard, and design an approximation algorithm to solve the problem with a theoretical bound of $1-\frac1 e $. We also devise some optimization techniques to speed up the solution. Extensive experiments on real-world datasets validate the effectiveness and the efficiency of our proposed solutions.

In spatial crowdsourcing, requesters submit their task-related locations and increase the demand of a local area. The platform prices these tasks and assigns spatial workers to serve if the prices are accepted by requesters. There exist mature pricing strategies which specialize in tackling the imbalance between supply and demand in a local market. However, in global optimization, the platform should consider the mobility of workers; that is, any single worker can be the potential supply for several areas, while it can only be the true supply of one area when assigned by the platform. The hardness lies in the uncertainty of the true supply of each area, hence the existing pricing strategies do not work. In the paper, we formally define this Global Dynamic Pricing(GDP) problem in spatial crowdsourcing. And since the objective is concerned with how the platform matches the supply to areas, we let the matching algorithm guide us how to price. We propose a MAtching-based Pricing Strategy (MAPS) with guaranteed bound. Extensive experiments conducted on the synthetic and real datasets demonstrate the effectiveness of MAPS.

There has been a dramatic growth of shared mobility applications such as ride-sharing, food delivery and crowdsourced parcel delivery. Shared mobility refers to transportation services that are shared among users, where a central issue is route planning . Given a set of workers and requests, route planning finds for each worker a route, i.e. , a sequence of locations to pick up and drop off passengers/parcels that arrive from time to time, with different optimization objectives. Previous studies lack practicability due to their conflicted objectives and inefficiency in inserting a new request into a route, a basic operation called insertion . In this paper, we present a unified formulation of route planning called URPSM. It has a well-defined parameterized objective function which eliminates the contradicted objectives in previous studies and enables flexible multi-objective route planning for shared mobility. We prove the problem is NP-hard and there is no polynomial-time algorithm with constant competitive ratio for the URPSM problem and its variants. In response, we devise an effective and efficient solution to address the URPSM problem approximately. We design a novel dynamic programming (DP) algorithm to accelerate the insertion operation from cubic or quadric time in previous work to only linear time. On basis of the DP algorithm, we propose a greedy based solution to the URPSM problem. Experimental results on real datasets show that our solution outperforms the state-of-the-arts by 1.2 to 12.8 times in effectiveness, and also runs 2.6 to 20.7 times faster.

Crowdsourcing has been shown to be effective in a wide range of applications, and is seeing increasing use. A large-scale crowdsourcing task often consists of thousands or millions of atomic tasks, each of which is usually a simple task such as binary choice or simple voting. To distribute a large-scale crowdsourcing task to limited crowd workers, a common practice is to pack a set of atomic tasks into a task bin and send to a crowd worker in a batch. It is challenging to decompose a large-scale crowdsourcing task and execute batches of atomic tasks, which ensures reliable answers at a minimal total cost. Large batches lead to unreliable answers of atomic tasks, while small batches incur unnecessary cost. In this paper, we investigate a general crowdsourcing task decomposition problem, called the Smart Large-scAle task DEcomposer (SLADE) problem, which aims to decompose a large-scale crowdsourcing task to achieve the desired reliability at a minimal cost. We prove the NP-hardness of the SLADE problem and propose solutions in both homogeneous and heterogeneous scenarios. For the homogeneous SLADE problem, where all the atomic tasks share the same reliability requirement, we propose a greedy heuristic algorithm and an efficient and effective approximation framework using an optimal priority queue (OPQ) structure with provable approximation ratio. For the heterogeneous SLADE problem, where the atomic tasks can have different reliability requirements, we extend the OPQ-based framework leveraging a partition strategy, and also prove its approximation guarantee. Finally, we verify the effectiveness and efficiency of the proposed solutions through extensive experiments on representative crowdsourcing platforms.

Taxi-calling apps are gaining increasing popularity for their efficiency in dispatching idle taxis to passengers in need. To precisely balance the supply and the demand of taxis, online taxicab platforms need to predict the Unit Original Taxi Demand (UOTD), which refers to the number of taxi-calling requirements submitted per unit time (e.g., every hour) and per unit region (e.g., each POI). Predicting UOTD is non-trivial for large-scale industrial online taxicab platforms because both accuracy and flexibility are essential. Complex non-linear models such as GBRT and deep learning are generally accurate, yet require labor-intensive model redesign after scenario changes (e.g., extra constraints due to new regulations). To accurately predict UOTD while remaining flexible to scenario changes, we propose LinUOTD, a unified linear regression model with more than 200 million dimensions of features. The simple model structure eliminates the need of repeated model redesign, while the high-dimensional features contribute to accurate UOTD prediction. We further design a series of optimization techniques for efficient model training and updating. Evaluations on two large-scale datasets from an industrial online taxicab platform verify that LinUOTD outperforms popular non-linear models in accuracy. We envision our experiences to adopt simple linear models with high-dimensional features in UOTD prediction as a pilot study and can shed insights upon other industrial large-scale spatio-temporal prediction problems.

The popularity of Online To Offline (O2O) service platforms has spurred the need for online task assignment in real-time spatial data, where streams of spatially distributed tasks and workers are matched in real time such that the total number of assigned pairs is maximized. Existing online task assignment models assume that each worker is either assigned a task immediately or waits for a subsequent task at a fixed location once she/he appears on the platform. Yet in practice a worker may actively move around rather than passively wait in place if no task is assigned. In this paper, we define a new problem Flexible Two-sided Online task Assignment (FTOA). FTOA aims to guide idle workers based on the prediction of tasks and workers so as to increase the total number of assigned worker-task pairs. To address the FTOA problem, we face two challenges: (i) How to generate guidance for idle workers based on the prediction of the spatiotemporal distribution of tasks and workers? (ii) How to leverage the guidance of workers' movements to optimize the online task assignment? To this end, we propose a novel two-step framework, which integrates offline prediction and online task assignment. Specifically, we estimate the distributions of tasks and workers per time slot and per unit area, and design an online task assignment algorithm, Prediction-oriented Online task Assignment in Real-time spatial data (POLAR-OP). It yields a 0.47-competitive ratio, which is nearly twice better than that of the state-of-the-art. POLAR-OP also reduces the time complexity to process each newly-arrived task/worker to O(1). We validate the effectiveness and efficiency of our methods via extensive experiments on both synthetic datasets and real-world datasets from a large-scale taxi-calling platform.