Dynamical networks are essential for understanding how network structures interact with dynamic processes over them. For example, in adaptive social networks, individuals' opinions influence and are influenced by their connections, leading to co-evolutionary patterns. Similarly, in neuroscience, the plasticity of neural networks dynamically reshapes their structure in response to activity. The topology of these networks profoundly impacts behavior, making their analysis critical in understanding stability, synchronization, or cascading failures.Uncertainty quantification (UQ) in dynamical networks addresses the challenges posed by incomplete or noisy knowledge of network structure, parameters, and external influences. For instance, fluctuating edge weights, evolving node connections, or stochastic interactions introduce uncertainties that affect predictions. In epidemiology, unknown contact patterns or varying transmission rates can significantly impact outbreak modeling, while in power systems, uncertainties in demand and renewable energy integration challenge reliability assessments. UQ systematically evaluates these uncertainties, offering techniques to quantify their effects and develop robust predictions. This 3-hour tutorial bridges the study of dynamical networks with the systematic framework of UQ, providing a comprehensive understanding of their interplay and practical applications.The tutorial begins with an introduction to dynamical networks, exploring their structural and behavioral characteristics through real-world examples in epidemiology, neuroscience, and engineering. It then transitions to UQ, covering foundational methods such as probabilistic simulations, sensitivity analysis, and stochastic modeling. Advanced topics, including machine learning-based surrogate modeling for computationally efficient UQ, will be discussed. The session concludes with an exploration of open challenges, such as integrating data-driven and physics-based models, and strategies for scaling UQ techniques to high-dimensional systems.

The Vesicular Stomatitis virus (VSV) causes cases of livestock disease that occur every year in regions in Mexico. Every few years, VSV spreads northwards into the U.S. in large outbreak events affecting hundreds of livestock premises across multiple states, leading to significant economic losses due to quarantines, trade restrictions, and veterinary expenses. VSV cases are mainly driven by biting arthropod vectors from multiple genera with different ecologies, making outbreak control challenging. The sporadic nature of outbreaks and limited understanding of transmission dynamics further hinder containment efforts, reducing the effectiveness of preemptive measures. In this paper, we propose an interpretable model to elucidate the key rules governing the spread of VSV. This model employs a sparse symbolic regression model, SINDy (Sparse Identification of Nonlinear Dynamical Systems), to identify the most significant ecological variables in spread dynamics, considering both spatial and temporal factors. Since many counties did not have VSV cases during the study period, counties were clustered into 40 regions incorporating static environmental variables land cover, soil properties, livestock density, and climate data and using spatially constrained Agglomerative Clustering based on geographic adjacency, resulting in an average region size of approximately 90 counties. Ecological variables included dynamic and static variables such as temperature, humidity, wind, soil characteristics, and altitude associated with vectors and hosts (cattle, horses, and mules). The change in cases from month to month by region was modeled using two SINDy variants: a baseline model with only ecological features (Normal) and an extended model incorporating spatially derived graph features (Graph).Each alpha was chosen to minimize CV-MSE while retaining less than 11 terms. Graphical features greatly reduced model error, and the SINDy model with select graphical features had a slightly better CV-MSE score than when all graphical features were included. All models identified the infected species as important in capturing the dynamics of case differences between regions.

Network interdiction problems, arising in critical applications from military strategy to disease control, involve a complex attacker-defender dynamic: one player optimizes a network-based objective, while the other strategically modifies the network to impede that objective. The inherent bi-level optimization and combinatorial nature of these problems pose a significant computational challenge, often rendering traditional exact solvers impractical and hindering the development of effective heuristics. While Graph Neural Networks (GNNs) have demonstrated promise in solving single-level combinatorial optimization problems on graphs, their direct application to bi-level interdiction problems remains limited. In this paper, we bridge this gap by introducing a novel approach that leverages the power of GNNs to learn Mixed-Integer Linear Programming (MILP) formulations of network interdiction problems. By representing the problem in this structured mathematical form, we empower a multipartite GNN with the representational capacity to effectively capture the complex interplay between the two players. This approach aligns the neural network with the underlying mathematical structure of interdiction problems, leading to improved performance. Through extensive experiments on two network interdiction tasks, we demonstrate the superiority of our proposed method over both baseline GNN models and traditional exact solvers, showcasing its potential for real-world applications.

Video understanding plays a fundamental role for content moderation on short video platforms, enabling the detection of inappropriate content. While classification remains the dominant approach for content moderation, it often struggles in scenarios requiring rapid and cost-efficient responses, such as trend adaptation and urgent escalations. To address this issue, we introduce an Embedding-Based Retrieval (EBR) method designed to complement traditional classification approaches. We first leverage a Supervised Contrastive Learning (SCL) framework to train a suite of foundation embedding models, including both single-modal and multi-modal architectures. Our models demonstrate superior performance over established contrastive learning methods such as CLIP and MoCo. Building on these embedding models, we design and implement the embedding-based retrieval system that integrates embedding generation and video retrieval to enable efficient and effective trend handling. Comprehensive offline experiments on 25 diverse emerging trends show that EBR improves ROC-AUC from 0.85 to 0.99 and PR-AUC from 0.35 to 0.95. Further online experiments reveal that EBR increases action rates by 10.32% and reduces operational costs by over 80%, while also enhancing interpretability and flexibility compared to classification-based solutions.

Influence maximization (IM) is the problem of identifying a limited number of initial influential users within a social network to maximize the number of influenced users. However, previous research has mostly focused on individual information propagation, neglecting the simultaneous and interactive dissemination of multiple information items. In reality, when users encounter a piece of information, such as a smartphone product, they often associate it with related products in their minds, such as earphones or computers from the same brand. Additionally, information platforms frequently recommend related content to users, amplifying this cascading effect and leading to multiplex influence diffusion. This paper first formulates the Multiplex Influence Maximization (Multi-IM) problem using multiplex diffusion models with an information association mechanism. In this problem, the seed set is a combination of influential users and information. To effectively manage the combinatorial complexity, we propose Graph Bayesian Optimization for Multi-IM (GBIM). The multiplex diffusion process is thoroughly investigated using a highly effective global kernelized attention message-passing module. This module, in conjunction with Bayesian linear regression (BLR), produces a scalable surrogate model. A data acquisition module incorporating the exploration-exploitation trade-off is developed to optimize the seed set further. Extensive experiments on synthetic and real-world datasets have proven our proposed framework effective. The code is available at https://github.com/zirui-yuan/GBIM.

Due to the significance of its various applications, source localization has garnered considerable attention as one of the most important means to confront diffusion hazards. Multi-source localization from a single-snapshot observation is especially relevant due to its prevalence. However, the inherent complexities of this problem, such as limited information, interactions among sources, and dependence on diffusion models, pose challenges to resolution. Current methods typically utilize heuristics and greedy selection, and they are usually bonded with one diffusion model. Consequently, their effectiveness is constrained. To address these limitations, we propose a simulation-based method termed BOSouL. Bayesian optimization (BO) is adopted to approximate the results for its sample efficiency. A surrogate function models uncertainty from the limited information. It takes sets of nodes as the input instead of individual nodes. BOSouL can incorporate any diffusion model in the data acquisition process through simulations. Empirical studies demonstrate that its performance is robust across graph structures and diffusion models. The code is available at https://github.com/XGraph-Team/BOSouL.

In recent years, Graph Neural Networks (GNNs) have attracted considerable attention. However, the rapid emergence of diverse GNN models, each grounded in different theoretical foundations, complicates the model selection process, as these models are not easily understood within a unified framework. Initial GNNs were constructed using spectral theory, while others were developed based on spatial theory. This theoretical divergence makes direct comparisons difficult. Furthermore, the variety of models within each theoretical domain further complicates their evaluation. In this tutorial, we explore state-of-the-art GNNs and present a comprehensive framework that bridges the spatial and spectral domains, clarifying their interrelationship. This framework deepens our understanding of GNN operations. The tutorial delves into key paradigms, such as spatial and spectral methods, through a synthesis of spectral graph theory and approximation theory. We conduct an in-depth analysis of recent research advancements, addressing emerging issues like over-smoothing, using well-established GNN models to illustrate the universality of our framework.

Influence Maximization (IM) is a critical area of research with widespread applications in viral marketing, social network recommendations, and disease containment. The primary objective of IM is to identify an optimal seed set that maximizes influence spread across networks. Traditional approaches to IM, including proxy-based, sketch-based, and simulation-based methods, each face specific limitations. Proxy-based methods often fail to capture complex seed interactions and are model-specific, sketch-based methods balance scalability with accuracy but can introduce errors, and simulation-based techniques, while accurate, are computationally intensive, particularly for large-scale graphs. Additionally, the relationship between seed set configurations and their resulting influence spreads remains largely a black box, posing significant challenges in modeling and prediction without extensive computational effort. To address these challenges, we introduce the Neural Tangent BOIM (NT-BOIM) framework, which utilizes Bayesian Optimization (BO) to reduce the number of required simulations significantly. This approach employs the Neural Tangent Kernel (NTK) as the kernel for the Gaussian Processes(GP) in our BO framework, enhancing our ability to model the complex, high-dimensional data typical of social networks. The NTK provides a robust framework to analyze and predict the training dynamics of neural networks, making it particularly effective for understanding and optimizing influence spread across different seed sets. Our NT-BOIM methodology not only enhances the performance of IM tasks but also expedites the optimization process, offering a computationally efficient alternative to traditional methods. Key innovations include designing a specialized NTK that accurately quantifies distances between seed sets in graph structures and implementing a stratified sampling technique, preceded by clustering, to ensure uniform sampling distribution within each BO iteration. Extensive empirical experiments demonstrate that our approach outperforms standard simulation methods in both effectiveness and computational speed, bridging the gap between computational efficiency and approximation accuracy. [Code] (https://github.com/XGraph-TeamlNT-BOIM).

Forecasting citations of scientific patents and publications is a crucial task for understanding the evolution and development of technological domains and for foresight into emerging technologies. By construing citations as a time series, the task can be cast into the domain of temporal point processes. Most existing work on forecasting with temporal point processes, both conventional and neural network-based, only performs single-step forecasting. In citation forecasting, however, the more salient goal is n -step forecasting: predicting the arrival of the next n citations. In this article, we propose Dynamic Multi-Context Attention Networks (DMA-Nets), a novel deep learning sequence-to-sequence (Seq2Seq) model with a novel hierarchical dynamic attention mechanism for long-term citation forecasting. Extensive experiments on two real-world datasets demonstrate that the proposed model learns better representations of conditional dependencies over historical sequences compared to state-of-the-art counterparts and thus achieves significant performance for citation predictions.

The US public school system is administered by local school districts. Each district comprises a set of schools mapped to attendance zones which are annually assessed to meet enrollment objectives. To support school officials in redrawing attendance boundaries, existing approaches have proven promising but still suffer from several challenges, including: 1) inability to scale to large school districts, 2) high computational cost of obtaining compact school attendance zones, and 3) lack of discussion on quantifying ethical considerations underlying the redrawing of school boundaries. Motivated by these challenges, this paper approaches the school redistricting problem from both computational and ethical standpoints. First, we introduce a practical framework based on sampling methods to solve school redistricting as a graph partitioning problem. Next, the advantages of adopting a modified objective function for optimizing discrete geometry to obtain compact boundaries are examined. Lastly, alternative metrics to address ethical considerations in real-world scenarios are formally defined and thoroughly discussed. Our findings highlight the inclusiveness and efficiency advantages of the designed framework and depict how tradeoffs need to be made to obtain qualitatively different school redistricting plans.

In this research, we propose an explanatory model that learns combinations of microbiome signatures that help predict reduced multidrug resistance (MDR) of disease-causing pathogens. A twin regression network in our model provides a means for estimating the MDR presence using microbiome samples. Further, we extend our model by using a backtracking approach that determines the effect of specific microbiome signatures on the prevalence of MDR, to identify potential beneficial combinations. Using high-throughput Illumina sequencing of the 16SrRNA gene data from 41 pastured poultry flocks in southeastern U.S. farms, the regression model identifies the factors influencing MDR prevalence in common foodborne pathogens (Salmonella, Campylobacter, and Listeria) at different stages of poultry growth.

This work develops a node-edge co-evolution model for attributed graph transformation, where both the node and edge attributes undergo changes due to complex interactions. Due to two fundamental obstacles, learning and approximating attributed graph transformation have not been thoroughly explored: 1) the difficulty of jointly considering four types of atomic interactions including nodes-to-edges, nodes-to-nodes, edges-to-nodes, and edges-to-edges interactions. 2) the difficulty of capturing iterative long-range interactions between nodes and edges. To solve these issues, we offer a novel and scalable equilibrium model, NEC∞, with node-edge message passing and edge-node message passing. Additionally, we propose an efficient optimization algorithm that is based on implicit gradient theorem and includes a theoretical analysis of NEC∞. The effectiveness and efficiency of the proposed model have been demonstrated through extensive experiments on synthetic and real-world data sets.

Weakly supervised referring expression grounding (WREG) is an attractive and challenging task for grounding target regions in images by understanding given referring expressions. WREG learns to ground target objects without the manual annotations between image regions and referring expressions during the model training phase. Different from the predominant grounding pattern of existing models, which locates target objects by reconstructing the region-expression correspondence, we investigate WREG from a novel perspective and enrich the prevailing pattern with self-knowledge distillation. Specifically, we propose a target-guided self-knowledge distillation approach that adopts the target prediction knowledge learned from the previous training iterations as the teacher to guide the subsequent training procedure. In order to avoid the misleading caused by the teacher knowledge with low prediction confidence, we present an uncertaintyaware knowledge refinement strategy to adaptively rectify the teacher knowledge by learning dynamic threshold values based on the model prediction uncertainty. To validate the proposed approach, we implement extensive experiments on three benchmark datasets, i.e., Ref Coco, RefCOCO+, and RefCOCOg. Our approach achieves new state-of-the-art results on several splits of the benchmark datasets, showcasing the advantage of the proposed framework for WREG. The implementation codes and trained models are available at: https://github.com/dami23IWREG.sar_KD.

Given its vast application on online social networks, Influence Maximization (IM) has garnered considerable attention over the last couple of decades. Due to the intricacy of IM, most current research concentrates on estimating the first-order contribution of the nodes to select a seed set, disregarding the higher-order interplay between different seeds. Consequently, the actual influence spread frequently deviates from expectations, and it remains unclear how the seed set quantitatively contributes to this deviation. To address this deficiency, this work dissects the influence exerted on individual seeds and their higher-order interactions utilizing the Sobol index, a variance-based sensitivity analysis. To adapt to IM contexts, seed selection is phrased as binary variables and split into distributions of varying orders. Based on our analysis with various Sobol indices, an IM algorithm dubbed SIM is proposed to improve the performance of current IM algorithms by over-selecting nodes followed by strategic pruning. A case study is carried out to demonstrate that the explanation of the impact effect can dependably identify the key higher-order interactions among seeds. SIM is empirically proved to be superior in effectiveness and competitive in efficiency by experiments on synthetic and real-world graphs.

Endoscopy has been routinely used to diagnose stomach diseases including intestinal metaplasia (IM) and gastritis atrophy (GA). Such routine examination usually demands highly skilled radiologists to focus on a single patient with substantial time, causing the following two key challenges: 1) the dependency on the radiologist's experience leading to inconsistent diagnosis results across different radiologists; 2) limited examination efficiency due to the demanding time and energy consumption to the radiologist. This paper proposes to address these two issues in endoscopy using novel machine learning method in three-folds. Firstly, we build a novel and relatively big endoscopy dataset of 21,420 images from the widely used White Light Imaging (WLI) endoscopy and more recent Linked Color Imaging (LCI) endoscopy, which were annotated by experienced radiologists and validated with biopsy results, presenting a benchmark dataset. Secondly, we propose a novel machine learning model inspired by the human visual system, named as local attention grouping, to effectively extract key visual features, which is further improved by learning from multiple randomly selected regional images via ensemble learning. Such a method avoids the significant problem in the deep learning methods that decrease the resolution of original images to reduce the size of input samples, which would remove smaller lesions in endoscopy images. Finally, we propose a dual transfer learning strategy to train the model with co-distributed features between WLI and LCI images to further improve the performance. The experiment results, measured by accuracy, specificity, sensitivity, positive detection rate and negative detection rate, on IM are 99.18 %, 98.90 %, 99.45 %, 99.45 %, 98.91 %, respectively, and on GA are 97.12 %, 95.34 %, 98.90 %, 98.86 %, 95.50 %, respectively, achieving state of the art performance that outperforms current mainstream deep learning models.

Predicting and quantifying the impact of traffic accidents is necessary and critical to Intelligent Transport Systems (ITS). As a state-of-the-art technique in graph learning, current graph neural networks heavily rely on graph Fourier transform, assuming homophily among the neighborhood. However, the homophily assumption makes it challenging to characterize abrupt signals such as traffic accidents. Our paper proposes an abrupt graph wavelet network (AGWN) to model traffic accidents and predict their time durations using only one single snapshot.

Chinese characters have semantic-rich compositional information in radical form. While almost all previous research has applied CNNs to extract this compositional information, our work utilizes deep graph learning on a compact, graph-based representation of Chinese characters. This allows us to exploit temporal information within the strict stroke order used in writing characters. Our results show that our stroke-based model has potential for helping large-scale language models on some Chinese natural language understanding tasks. In particular, we demonstrate that our graph model produces more interpretable embeddings shown through word subtraction analogies and character embedding visualizations.

Influence blocking maximization (IBM) is crucial in many critical real-world problems such as rumors prevention and epidemic containment. The existing work suffers from: (1) concentrating on uniform costs at the individual level, (2) mostly utilizing greedy approaches to approximate optimization, (3) lacking a proper graph representation for influence estimates. To address these issues, this research introduces a neural network model dubbed Neural Influence Blocking (\algo) for improved approximation and enhanced influence blocking effectiveness. The code is available at https://github.com/oates9895/NIB.

Many real-world networks evolve over time, which results in dynamic graphs such as human mobility networks and brain networks. Usually, the “dynamics on graphs” (e.g., node attribute values evolving) are observable, and may be related to and indicative of the underlying “dynamics of graphs” (e.g., evolving of the graph topology). Traditional RNN-based methods are not adaptive or scalable for learn- ing the unknown mappings between two types of dynamic graph data. This study presents a AD-ESN, and adaptive echo state network that can automatically learn the best neural net- work architecture for certain data while keeping the efficiency advantage of echo state networks. We show that AD-ESN can successfully discover the underlying pre-defined map- ping function and unknown nonlinear map-ping between time series and graphs.

Modeling the spatiotemporal nature of the spread of infectious diseases can provide useful intuition in understanding the time-varying aspect of the disease spread and the underlying complex spatial dependency observed in people's mobility patterns. Besides, the county level multiple related time series information can be leveraged to make a forecast on an individual time series. Adding to this challenge is the fact that real-time data often deviates from the unimodal Gaussian distribution assumption and may show some complex mixed patterns. Motivated by this, we develop a deep learning-based time-series model for probabilistic forecasting called Auto-regressive Mixed Density Dynamic Diffusion Network (ARM3Dnet), which considers both people's mobility and disease spread as a diffusion process on a dynamic directed graph. The Gaussian Mixture Model layer is implemented to consider the multimodal nature of the realtime data while learning from multiple related time series. We show that our model, when trained with the best combination of dynamic covariate features and mixture components, can outperform both traditional statistical and deep learning models in forecasting the number of Covid-19 deaths and cases at the county level in the United States.

Predicting the impact of traffic i ncidents b ased on traffic s ensor d ata i s a n e ssential r esearch t opic i n t he fi eld of Intelligent Transportation Systems (ITS). Tackling the problem of estimating the durations of incidents from their early stages is a challenge due to the variable nature of such incidents and the complex structure of modern road networks. Existing studies on forecasting the incident duration from sensor data are mostly incapable of modeling 1) the spatiotemporal correlations of traffic s ensors a nd a rterial r oads a nd 2 ) t he hierarchical topology of the traffic sensor and road networks. In this paper, we propose the Hierarchical Attention-based Spatiotemporal Graph Convolutional Network model (HastGCN) to solve the incident duration forecasting problem by formulating the spatiotemporal correlation and traffic p atterns o n b oth t he s ensor l evel and the road level in their natural hierarchical manner. At the sensor level, we propose a spatiotemporal attention mechanism followed by graph convolutions to model the local correlations and patterns between traffic s ensors o n t he s ame a rterial road. At the road level, a connectivity-aware attention mechanism is designed to learn the global spatial relatedness between each arterial road. Traffic-condition a ware g raph c onvolutions are then applied to understand the target incident representation for the incident duration forecasting.

Uplink transmission delay has been considered as a main component of the end-to-end delay in the fog radio access networks (F-RAN). However, existing analysis of the transmission delay ignore the packet retransmission delay, i.e., mean local delay (MLD), which is the main component of transmission delay. In addition, the MLD has not been investigated in a clustered F-RAN. Therefore, in this work, we leverage the Matern cluster process (MCP) to analyse MLD in a large-scale F-RAN. To derive the MLD, we obtain the uplink coverage probability (CP) firstly. The MLD can be derived by the moment result of the CP, which is defined as the probability of received signal-to-interference-ratio (SIR) is larger than a threshold. To obtain the moment result of the CP, the analytical result of the CP and its approximation with a lower computational complexity are derived. However, the moment result of CP is difficult to be validated via simulations. Therefore, we derive the meta distribution, which is calculated directly from the moment result of the CP, and validate its correctness by Monte Carlo simulations. Equipped with this, the analytical results of MLD in closed-form are derived. Based on these results, the effect of UE activity factor and the uplink power control (PC) factor on the MLD are analysed. The results show that the fog access points (FAP) density has no effect on the MLD.

Public school boundaries are redrawn from time to time to ensure effective functioning of school systems. This process, also called school redistricting, is non-trivial due to (1) the presence of multiple design criteria such as capacity utilization, proximity and travel time which are hard for planners to consider simultaneously, (2) the fixed locations of schools with widely differing capacities that need to be balanced, (3) the spatial nature of the data and the need to preserve contiguity in school zones, and (4) the difficulty in quantifying local factors that may arise. Motivated by these challenges and the intricacy of the process, we propose a geospatial clustering algorithm called GeoKmeans for assisting planners in designing school boundaries such that students are assigned to proximal schools while ensuring effective utilization of school capacities. The algorithm operates on polygonal geometries and connects them into geographically contiguous school boundaries while balancing problem-specific constraints. We evaluate our approach on real-world data of two rapidly growing school districts in the US. Results indicate the efficacy of our approach in designing boundaries. Additionally, a case study is included to demonstrate the potential of GeoKmeans to assist planners in drawing boundaries.

Circuit obfuscation has been proposed to protect digital integrated circuits (ICs) from different security threats such as reverse engineering by introducing ambiguity in the circuit, i.e., the addition of the logic gates whose functionality cannot be determined easily by the attacker. In order to conquer such defenses, techniques such as Boolean satisfiability-checking (SAT)-based attacks were introduced. SAT-attack can potentially decrypt the obfuscated circuits. However, the deobfuscation runtime could have a large span ranging from few milliseconds to a few years or more, depending on the number and location of obfuscated gates, the topology of the obfuscated circuit and obfuscation technique used. To ensure the security of the deployed obfuscation mechanism, it is essential to accurately pre-estimate the deobfuscation time. Thereby one can optimize the deployed defense in order to maximize the deobfuscation runtime. However, estimating the deobfuscation runtime is a challenging task due to 1) the complexity and heterogeneity of the graph-structured circuit, 2) the unknown and sophisticated mechanisms of the attackers for deobfuscation, 3) efficiency and scalability requirement in practice. To address the challenges mentioned above, this work proposes the first machine-learning framework that predicts the deobfuscation runtime based on graph deep learning. Specifically, we design a new model, ICNet with new input and convolution layers to characterize the circuit's topology, which is then integrated by composite deep fully-connected layers to obtain the deobfuscation runtime. The proposed ICNet is an end-to-end framework that can automatically extract the deter-minant features required for deobfuscation runtime prediction. Extensive experiments on standard benchmarks demonstrate its effectiveness and efficiency beyond many competitive baselines.

Jianfeng He, Xuchao Zhang, Shuo Lei, Zhiqian Chen, Fanglan Chen, Abdulaziz Alhamadani, Bei Xiao, ChangTien Lu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

Spatial optimization problems (SOPs) are characterized by spatial relationships governing the decision variables, objectives and/or constraint functions. These are mostly combinatorial problems (NP-hard) due to the presence of discrete spatial units. Hence, exact optimization methods cannot solve them optimally under practical time constraints, especially for large-sized instances. Motivated by this challenge, we explore the use of population-based metaheuristics for solving SOPs. To this end, we observe that the search moves employed by these methods are suited to real-parameter continuous search space rather. To adapt them to the SOPs, we explore the role of domain knowledge in designing spatially-aware search operators that can efficiently search for an optimal solution in discrete search space while respecting the spatial constraints. These modifications result in a simple yet highly effective spatial hybrid metaheuristic called SPATIAL, which is applied to the problem of school boundary formation (also called school redistricting). Experimental findings on real-world datasets reveal the efficacy of our algorithm in obtaining superior quality solutions in comparison to traditional baseline methods. Additionally, we perform an in-depth study of the individual components of our framework and highlight the flexibility of our method in assimilating other search operators as well as in adapting it to related SOPs.

Traffic prediction is a challenging task due to the time-varying nature of traffic patterns and the complex spatial dependency of road networks. Adding to the challenge, there are a number of errors introduced in traffic sensor reporting, including bias and noise. However, most of the previous works treat the sensor observations as exact measures ignoring the effect of unknown noise. To model the spatial and temporal dependencies, existing studies combine graph neural networks (GNNs) with other deep learning techniques but their equal weighting of different dependencies limits the models' ability to capture the real dynamics in the traffic network. To deal with the above issues, we propose a novel deep learning framework called Deep Kalman Filtering Network (DKFN) to forecast the network-wide traffic state by modeling the self and neighbor dependencies as two streams, and their predictions are fused under the statistical theory and optimized through the Kalman filtering network. First, the reliability of each stream is evaluated using variances. Then, the Kalman filter is leveraged to properly fuse noisy observations in terms of their reliability. Experimental results reflect the superiority of the proposed method over baseline models on two real-world traffic datasets in the speed prediction task.

Due to constant shifts in population and changing demographics, school boundary processes take place to make adjustments to school attendance zones. This spatial problem has multiple criteria like locations of schools, their capacity utilization, proximity, presence of geographical/ man-made barriers, etc. In this paper, we formulate the problem of designing school boundaries as a spatially-constrained clustering/ regionalization problem and propose an automated approach called REGAL for solving it. REGAL is two-stage framework that starts by creating a candidate solution with regard to domain constraints such as school locations and spatial contiguity. Then a local search method improves the quality of the candidate solution by optimizing population balance and compactness of school zones while satisfying problem constraints. Experimentally, we demonstrate the efficacy of the REGAL framework on actual datasets from two school districts in the US.

Modeling and forecasting forward citations to a patent is a central task for the discovery of emerging technologies and for measuring the pulse of inventive progress. Conventional methods for forecasting these forward citations cast the problem as analysis of temporal point processes which rely on the conditional intensity of previously received citations. Recent approaches model the conditional intensity as a chain of recurrent neural networks to capture memory dependency in hopes of reducing the restrictions of the parametric form of the intensity function. For the problem of patent citations, we observe that forecasting a patent's chain of citations benefits from not only the patent's history itself but also from the historical citations of assignees and inventors associated with that patent. In this paper, we propose a sequence-to-sequence model which employs an attention-of-attention mechanism to capture the dependencies of these multiple time sequences. Furthermore, the proposed model is able to forecast both the timestamp and the category of a patent's next citation. Extensive experiments on a large patent citation dataset collected from USPTO demonstrate that the proposed model outperforms state-of-the-art models at forward citation forecasting.

One can infer from the broken window theory that the perception of a city street's safety level relies significantly on the visual appearance of the street. Previous works have addressed the feasibility of using computer vision algorithms to classify urban scenes. Most of the existing urban perception predictions focus on binary outcomes such as safe or dangerous, wealthy or poor. However, binary predictions are not representative and cannot provide informative inferences such as the potential crime types in certain areas. In this paper, we explore the connection between urban perception and crime inferences. We propose a convolutional neural network (CNN) - StreetNet to learn crime rankings from street view images. The learning process is formulated on the basis of preference learning and label ranking settings. We design a street view images retrieval algorithm to improve the representation of urban perception. A data-driven, spatiotemporal algorithm is proposed to find unbiased label mappings between the street view images and the crime ranking records. Extensive evaluations conducted on images from different cities and comparisons with baselines demonstrate the effectiveness of our proposed method.

For node level graph encoding, a recent important state-of-art method is the graph convolutional networks (GCN), which nicely integrate local vertex features and graph topology in the spectral domain. However, current studies suffer from several drawbacks: (1) graph CNNs rely on Chebyshev polynomial approximation which results in oscillatory approximation at jump discontinuities; (2) Increasing the order of Chebyshev polynomial can reduce the oscillations issue, but also incurs unaffordable computational cost; (3) Chebyshev polynomials require degree Ω(poly(1/ε)) to approximate a jump signal such as |x|, while rational function only needs O(poly log(1/ε)). However, it is non-trivial to apply rational approximation without increasing computational complexity due to the denominator. In this paper, the superiority of rational approximation is exploited for graph signal recovering. RatioanlNet is proposed to integrate rational function and neural networks. We show that the rational function of eigenvalues can be rewritten as a function of graph Laplacian, which can avoid multiplication by the eigenvector matrix. Focusing on the analysis of approximation on graph convolution operation, a graph signal regression task is formulated. Under graph signal regression task, its time complexity can be significantly reduced by graph Fourier transform. To overcome the local minimum problem of neural networks model, a relaxed Remez algorithm is utilized to initialize the weight parameters. Convergence rate of RatioanlNet and polynomial based methods on a jump signal is analyzed for a theoretical guarantee. The extensive experimental results demonstrated that our approach could effectively characterize the jump discontinuities, outperforming competing methods by a substantial margin on both synthetic and real-world graphs.

Self-paced learning (SPL) mimics the cognitive process of humans, who generally learn from easy samples to hard ones. One key issue in SPL is the training process required for each instance weight depends on the other samples and thus cannot easily be run in a distributed manner in a large-scale dataset. In this paper, we reformulate the self-paced learning problem into a distributed setting and propose a novel Distributed Self-Paced Learning method (DSPL) to handle large scale datasets. Specifically, both the model and instance weights can be optimized in parallel for each batch based on a consensus alternating direction method of multipliers. We also prove the convergence of our algorithm under mild conditions. Extensive experiments on both synthetic and real datasets demonstrate that our approach is superior to those of existing methods.

The scarcity of potable water is a critical challenge in many regions around the world. Previous studies have shown that knowledge of device-level water usage can lead to significant conservation. Although there is considerable interest in determining discriminative features via sparse coding for water disaggregation to separate whole-house consumption into its component appliances, existing methods lack a mechanism for fitting coefficient distributions and are thus unable to accurately discriminate parallel devices’ consumption. This article proposes a Bayesian discriminative sparse coding model, referred to as Virtual Metering (VM), for this disaggregation task. Mixture-of-Gammas is employed for the prior distribution of coefficients, contributing two benefits: (i) guaranteeing the coefficients’ sparseness and non-negativity, and (ii) capturing the distribution of active coefficients. The resulting method effectively adapts the bases to aggregated consumption to facilitate discriminative learning in the proposed model, and devices’ shape features are formalized and incorporated into Bayesian sparse coding to direct the learning of basis functions. Compact Gibbs Sampling (CGS) is developed to accelerate the inference process by utilizing the sparse structure of coefficients. The empirical results obtained from applying the new model to large-scale real and synthetic datasets revealed that VM significantly outperformed the benchmark methods.

In the era of information overload, people are struggling to make sense of complex story events in massive social media data. Most existing approaches are designed to address event extraction in news reports, documents and abstracts, but such approaches are not suitable for Twitter data streams due to their unstructured language, short-length messages, and heterogeneous features; few existing approach generates a story by considering both the shared topics throughout the story and the smooth connection between successive nodes simultaneously. In this paper, a novel Twitter stoRy generation framework via shAred subspaCe and tEmporal Smoothness called TRACES is proposed. Given a query of an ongoing event, a novel multi-task clustering method integrated with shared subspace and temporal smoothness (STMTC) is proposed to generate the event stories. Extensive experimental evaluations of data sets for different events demonstrate the effectiveness of this new approach.

In today's era of information overload, people are struggling to detect the evolution of hot topics from massive news media and microblogs such as Twitter. Reports from mainstream news agencies and discussions from microblogs could complement each other to form a complete picture of major events. Existing work has generally focused on a single source, seldom attempting to combine multiple sources to track the evolution of topics: emerging, evolving and fading phrases as this would require a considerably more sophisticated model. This paper proposes a novel story discovery model that integrates evolutionary topics in news and Twitter data sources using an incremental algorithm by 1) discovering complementary information from news and microblogs that provides a more complete view of major events; 2) modeling emerging, evolving and fading topics and features throughout ongoing events; and 3) creating a scalable algorithm that is capable of handling massive data from news and social media. The parameters of the new model are optimized using a novel algorithm based on the alternative direction method of multipliers (ADMM). Extensive experimental evaluations on multiple datasets from different domains demonstrate the effectiveness and efficiency of our proposed approach.

FusionGAN is a novel genre fusion framework for music generation that integrates the strengths of generative adversarial networks and dual learning. In particular, the proposed method offers a dual learning extension that can effectively integrate the styles of the given domains. To efficiently quantify the difference among diverse domains and avoid the vanishing gradient issue, FusionGAN provides a Wasserstein based metric to approximate the distance between the target domain and the existing domains. Adopting the Wasserstein distance, a new domain is created by combining the patterns of the existing domains using adversarial learning. Experimental results on public music datasets demonstrated that our approach could effectively merge two genres.

Deriving event storylines is an effective summarization method to succinctly organize extensive information, which can significantly alleviate the pain of information overload. The critical challenge is the lack of widely recognized definition of storyline metric. Prior studies have developed various approaches based on different assumptions about users' interests. These works can extract interesting patterns, but their assumptions do not guarantee that the derived patterns will match users' preference. On the other hand, their exclusiveness of single modality source misses cross-modality information. This paper proposes a method, multimodal imitation learning via Generative Adversarial Networks(MIL-GAN), to directly model users' interests as reflected by various data. In particular, the proposed model addresses the critical challenge by imitating users' demonstrated storylines. Our proposed model is designed to learn the reward patterns given user-provided storylines and then applies the learned policy to unseen data. The proposed approach is demonstrated to be capable of acquiring the user's implicit intent and outperforming competing methods by a substantial margin with a user study.

Connecting the dots between diverse entities such as people and organizations is a vital task for forming hypotheses and uncovering latent relationships among complex and large datasets. Most existing approaches are designed to address the relationship of entities in news reports, documents and abstracts, but such approaches are not suitable for Twitter data streams due to their unstructured languages, short-length messages, heterogeneous features and massive size. The sheer size of Twitter data requires more efficient algorithms to connect the dots within a short period of time. We present a system that automatically constructs stories by connecting entities in Twitter datasets. An entity similarity model is designed that combines both traditional entity-related features and social network attributes and a novel story generation algorithm applied on the similarity model is proposed to cope with the massive Twitter datasets. Extensive experimental evaluations were conducted to demonstrate the effectiveness of this new approach.

Numerous web crawlers especially impolite crawlers visit websites to get contents every day, which yields higher access frequency than the websites can hold. The big traffic of impolite crawlers causes a strong hazard on analysis of normal users and advertisement income. In this paper, we present a method to detect impolite crawlers by using time series analysis. This method is applied to real data of web server logs. Compared with the old methods only using common log attributes as features, the method using time series features improves detection accuracy by at least 20%.