Artificial intelligence (AI) models are increasingly deployed directly in web browsers to enable low-latency, privacy-preserving inference. While this shift offers significant usability and scalability benefits, it also exposes model code and parameters to untrusted environments, leaving them vulnerable to theft, reverse engineering, and tampering. Our analysis demonstrates that existing JavaScript-based inference frameworks are highly susceptible to model extraction, posing serious security and intellectual property risks. To address this gap, we present WAMO, a WebAssembly-based obfuscation framework that secures browser-side AI models. WAMO introduces a comprehensive conversion pipeline that translates mainstream model formats into Wasm-native modules, applying model-specific obfuscation at the Wasm layer to target weights, operators, and computation graphs. This design shifts model execution from easily inspected JavaScript assets to hardened Wasm binaries, significantly raising the difficulty of static and dynamic analysis. Evaluation shows that WAMO increases cyclomatic complexity by 71.0% and Halstead effort by 455.57%, while incurring < 1% accuracy loss and no inference slowdown.

Control flow errors (CFEs) pose a serious threat to the reliability of embedded systems, particularly under increasing integration density and shrinking feature sizes. Existing CFE detection techniques typically rely on coarse-grained analysis and uniform checking strategies, lacking fine-grained awareness of structural and runtime characteristics. This limitation often leads to considerable overhead, making such approaches less suitable for resource-constrained embedded systems. To tackle this shortcoming, we propose a CFE Detection approach guided by Basic block Vulnerability Analysis (CDBVA) that aims to strike a balance between the detection effectiveness and the overhead. Specifically, we first extract the CFE-related structural and execution features to characterize basic block vulnerability. Then, we train a learning-based model to predict basic blocks that are vulnerable to CFEs. Finally, we design a hybrid signature checking strategy that performs appropriate checks on vulnerable and non-vulnerable basic blocks separately. Experimental results demonstrate that CDBVA achieves an average prediction accuracy of 86.2% and an average CFE coverage of 95.79%, outperforming state-of-the-art approaches. While maintaining high CFE coverage, CDBVA improves the evaluation factor by 12.14%–20.82%, achieving a favorable tradeoff between detection effectiveness and overhead. In addition, CDBVA demonstrates stable performance across diverse input conditions and heterogeneous hardware architectures.

We present SPASCA - a conversational AI system that promotes psychological and cognitive well-being of persons living with dementia (PLWD). This system features an AI agent that provides social presence and support to PLWD through verbal communications, without physical presence of human caregivers. The system integrates (1) a novel dialogue model that generates dialogue items relevant to the user's experiences and lifestyle, (2) a digital avatar in the form of a talking head with the identity of a caregiver who is familiar to the demented user. We develop prototypes that adopt various interaction modalities and conversational styles and report the pros and cons of different system configurations through expert review. Our system shows the potential of conversational AI for personalized and affordable healthcare services.

Decompilation is one of the key techniques in software reverse engineering and is widely used in security-related tasks such as malware analysis and vulnerability detection. Software is typically released in binary form with symbol information stripped. Although decompilers are capable of reconstructing a lot of the information lost during compilation, they often fail to recover meaningful variable names. As a result, the readability of the decompiled code is significantly reduced, severely hindering the efficiency of reverse analysis. To address this problem, we propose a novel variable name recovery approach ReDevar for the decompiled code based on Multi-task Learning (MTL), which takes variable name recovery as the main task and introduces two additional tasks, including data source prediction and name complexity prediction, corresponding to two aspects of semantic understanding and semantic transformation of variables in the decompiled code. Data source prediction simulates the data flow relationship among variable instances, assisting ReDevar better understand the semantics of variables. The name complexity prediction task enables ReDevar to perceive the complexity of the expected variable name at a mask position, obtaining appropriate semantic transformation results. They are both jointly trained with the variable name recovery task to improve the understanding of ReDevar for variable semantics and name composition. We conducted a series of experiments to validate the effectiveness of ReDevar. Experimental results show that ReDevar achieves top- 1 accuracy of $54.44 \%$ on the VarCorpus benchmark dataset, 2.81% and 1.50% higher than the state-of-the-art approaches VarBERT and Resym respectively. In addition, we also evaluated the performance of ReDevar under various conditions, including different dataset splitting strategies, different decompilers, and different optimization levels. The results demonstrate that ReDevar generalizes well across all settings. Furthermore, the ablation study indicates that the two auxiliary tasks we introduced in ReDevar are both beneficial for the variable name recovery task.

Secure Aggregation (SA) is a cornerstone of Federated Learning (FL), ensuring that user updates remain hidden from servers. The advanced Flamingo (S&P’23) has realized multi-round aggregation and improved efficiency. However, it still faces several key challenges: scalability issues with dynamic user participation, a lack of verifiability for server-side aggregation results, and vulnerability to Model Inconsistency Attacks (MIA) caused by a malicious server distributing inconsistent models. To address these issues, we propose $\textit{Janus}$, a generic SA scheme based on dual-server architecture. Janus ensures security against up to $n-2$ colluding clients (where $n$ is the total client count), which prevents privacy breaches for non-colluders. Additionally, Janus is model-independent, ensuring applicability across any FL model without specific adaptations. Furthermore, Janus introduces a new cryptographic primitive, Separable Homomorphic Commitment, which enables clients to efficiently verify the correctness of aggregation. Finally, extensive experiments show that Janus not only significantly enhances security but also reduces per-client communication and computation overhead from logarithmic to constant scale, with a tolerable impact on model performance.

Discovering potential vulnerabilities has long been a fundamental goal in software security. Among them, bit flips, caused by hardware or environmental disturbances, are increasingly recognized as a new type of vulnerabilities that threaten program reliability at the instruction level. However, existing work is often restricted to individual programs and requires retraining when applied to unseen code, severely limiting their practicality and responsiveness. In this paper, we propose CIVP, a novel framework for context-aware instruction vulnerability prediction, generalizing to unseen programs without retraining. Specifically, to capture the rich contextual semantics of instructions, CIVP first leverages Large Language Models (LLMs) to accurately extract semantic embeddings of instructions. Then, CIVP further constructs an instruction execution graph containing complex relations of program execution, which implicates the potential path of error propagation. To improve instruction representation for vulnerability prediction, CIVP enhances GraphSAGE with multi-hop diffusion to capture inter-program structural patterns and contextual dependencies, and adopts pseudo-labeling to improve the model’s generalization for vulnerable instructions. Extensive experiments on a dataset of 26 real-world programs demonstrate that CIVP significantly outperforms the state-of-the-art approaches, achieving up to 20.5%↑ Recall and 18.5%↑ F1-score improvements. Notably, CIVP generalizes well to unseen programs, offering an efficient and scalable solution for proactive instruction-level hardening before software deployment.

As GPUs are widely deployed in High Performance Computing systems, it is critical to ensure that these systems can perform reliably. To improve system reliability, researchers estimate the error resilience of GPU programs by understanding resilience characteristics or modeling error propagation. However, features indicative of resilience rely on manual extraction from simulations of numerous faults, and error propagation analysis cannot target fine-grained bit-level faults. To address those problems, this paper introduces a novel paradigm, namely InstrDGM, for efficiently predicting GPU error resilience. Specifically, InstrDGM first fine-tunes a large language model using extensive sequences of GPU assembly instructions for extracting the semantic representation of instructions automatically. Meanwhile, we consider the propagation of bit-level faults during instruction execution and data transfer processes, and leverage graph neural networks to capture their distinct error propagation patterns. Then, the fault embeddings extracted from these error propagation patterns are integrated for error resilience prediction. Additionally, this paper releases a new dataset for GPU error resilience assessment, containing 1.2 million fault samples. Finally, extensive experiments show that InstrDGM significantly outperforms existing methods.

The exponential growth of Large Language Model (LLM) training demands in HPC systems has exposed critical reliability challenges, particularly from transient faults. Unlike resilience studies in conventional DNN inference, the massive parameter scale and iterative updates in LLM training trigger more complex failure patterns. To address these challenges, we introduce LLMFI, a new fault injection tool, and reveal six distinct failure behaviors through 300K+ fault injection experiments (exceeding 5K GPU node-hours). Our key insight is that, while most injected faults are eventually masked by the training iteration mechanism, a critical subset leads to catastrophic failures or performance degradation. Further, we propose LLMFT, a novel machine-learning-based fault tolerance framework that implements closed-loop error control via heuristic feature extraction, fault detector, and dual recovery mechanisms. Extensive evaluation demonstrates that LLMFT achieves an average of 97.61% F1-score in fault detection with only 0.01%–0.05% additional GPU memory overhead, effectively mitigating LLM training failures.

Despite the vital role of recommendation systems (RS) in delivering personalized services tailored to users' needs, user fairness issues have increasingly emerged in recent years, especially differentiated treatments caused by user sensitive attributes. This not only undermines both user experience and platform revenues, but also leads to potential social unfairness. Although many fairness-aware methods have been developed and achieved some success, many of them filter out sensitive attribute information while ignoring the potential loss of personalized information, leading to suboptimal results. Large language models (LLMs) have demonstrated remarkable capabilities across various tasks, while their potential in fairness-aware recommendation remains further unexplored. In this paper, we propose a new exploration of fairness-aware RS by prompting LLMs with the user's personalized fairness degrees to augment fair user-item interaction for training. Specifically, to estimate the fairness degree of each user, we first design a personalized unfairness modelling module, consisting of a replaceable fairness-aware representation learning model. Moreover, to enable LLMs to perceive fairness from semantic information and adapt to various scenarios, we propose a prompt tuning mechanism to optimize user-shared prompt templates with the objective of maximizing the consistency with users' preferences and the diversity of augmented data. Finally, we utilize LLMs to augment fair interaction data with the optimal prompts and integrate it with the raw data to re-train the recommendation model. Extensive experiments on two real-world datasets demonstrate the superiority of our approach in terms of recommendation performance, fairness, and robustness.

WebAssembly (Wasm) is a universal low-level bytecode designed to build modern web systems. Recent studies have shown that technologies such as voltage scaling and RowHammer attacks are expected to increase the likelihood of bit flips, which may cause unacceptable or catastrophic system failures. This raises concerns about the impact of bit flips on Wasm programs, which run as instructions in web systems, and it is an undeveloped topic since the features of Wasm differ from traditional programs. In this paper, we propose a novel paradigm, namely IVPSEG, to understand the error propagation of bit flips within Wasm programs. Specifically, we first use Large Language Models (LLMs) to automatically extract instruction embeddings containing semantic knowledge of each instruction's context. Then, we exploit these embeddings and program structure (control execution and data transfer) to construct a semantic enhanced code property graph, which implicates the potential path of error propagation. Based on this graph, we utilize graph neural networks and attention diffusion to optimize instruction embeddings by capturing different error propagation patterns for instruction vulnerability prediction. In particular, we build a Wasm compilation and fault generation system to simulate bit flips at Wasm runtime. Our experimental results with 14 benchmark programs and test cases show IVPSEG outperforms the state-of-the-art methods in terms of accuracy (average 13.06%ͽ ), F1-score (average 14.93%↑), and model robustness.

Federated learning (FL) in Vehicular Ad Hoc Networks (VANETs) enables vehicles to collaboratively train a global model for intelligent transportation systems while preserving the privacy of their local data. However, the openness and dynamic nature of VANETs introduce significant security challenges, including identity privacy leakage, model inversion attacks, and compromised model integrity. Existing cryptographic solutions, such as differential privacy and homomorphic encryption, provide partial mitigation but suffer from drawbacks including inefficiency, limited data utility, and vulnerability to data poisoning attacks. To tackle these challenges, this paper introduces FedLG, a generic certificateless (CL) authentication scheme with conditional anonymity. FedLG ensures trustworthy FL by integrating multiple security mechanisms that together guarantee model authenticity, data integrity, and privacy. Specifically, FedLG leverages Type-T signatures as a blackbox to ensure the authenticity and integrity of model parameters shared by anonymous vehicles. Additionally, we introduce a novel public key reconstruction mechanism to enhance the security of traditional CL-based systems, effectively mitigating common public key replacement attacks. FedLG also incorporates batch verification with an adaptive group batch verification algorithm, dynamically adjusting batch sizes to identify invalid signatures while preserving valid data, thereby facilitating faster model convergence. Moreover, FedLG maintains the utility of user-contributed data and can seamlessly integrate it with data poisoning attack prevention mechanisms to enhance security further. Experimental results show that FedLG is model-independent, as its integration does not affect the original model’s performance on its dataset. Moreover, it reduces the computational overhead of signature generation and verification by at least 30.8% and 56.3%, respectively, achieving an overall efficiency improvement of 49.69% compared to state-of-the-art FL authentication protocols for VANETs.

The demand for more precise and timely urban resource allocation and management has driven the extension of urban flow prediction from short-term to long-term horizons. As the time scale expands, the issue of urban flow distribution shift becomes increasingly prominent due to various impact factors, such as weather, events, city changes, etc. Traditionally, comprehensively analyzing and addressing the causal relationships underlying the distribution shift caused by these factors has been challenging. In this paper, we propose that these impact factors can be partitioned in two major types, i.e., context factors and structural factors. We then present a decomposition-based model for long-term urban flow prediction from a causal perspective, namedDeCau, which can discriminate between the two types of factors for effectively solving the problem of urban flow distribution shift. First, we employ a decomposition module to decompose urban flow into seasonal part and trend part. The seasonal part contains high frequency irregular variations caused by context factors. We advise a shared distribution estimator to approximate the unavailable prior distributions of context factors, and then apply causal intervention to mitigate the confounding impact of context factors. The distribution shift in the trend part is induced by structural factors. We design a dual causal dependency extractor to model the causality between POIs distribution and urban flow, and then eliminate spurious correlations through causal adjustment. Finally, we design an end-to-end framework for long-term urban flow prediction by combining the embeddings from two parts, enabling the model to generalize to unseen distribution. Extensive experimental results demonstrateDeCauoutperforms state-of-the-art baselines.

Control flow errors (CFEs) are prevalent and de-structive runtime faults that compromise software reliability and security. Existing CFE detection approaches either perform coarse-grained modeling at the basic-block level, missing fine-grained implicit features, or rely on exhaustive fault injection for detailed error patterns. These limitations hinder efficient detection, especially for silent CFEs, where branches deviate from correct paths, yet still follow compiler-defined ones. To bridge this gap, we propose CEDAR, a novel CFE detection approach, which detects silent CFEs by localizing and hardening vulnerable instructions. Specifically, we first conduct multi-level control flow analysis with limited fault injection, enhanced by dynamic data propagation, to construct a dual-layer heterogeneous graph representation of the program. Subsequently, we employ Graph Neural Networks (GNNs) to learn silent CFE-relevant embeddings and develop a model to localize vulnerable instructions. Finally, targeted program hardening mechanisms are integrated to detect silent CFEs and improve overall CFE coverage. Experiments demonstrate that CEDAR achieves 92.28% coverage of silent CFEs and 96.47% on average for overall CFEs. While maintaining high coverage, it improves the Evaluation Factor (EF) by 21.97% over the state-of-the-art approach, achieving a favorable trade-off between effectiveness and overhead. More-over, it demonstrates robustness across diverse input scenarios and Instruction Set Architectures (ISAs).

Origin-destination (OD) crowd flow, if more accurately inferred at a fine-grained level, has the potential to enhance the efficacy of various urban applications. While in practice for mining OD crowd flow with effect, the problem of spatially interpolating OD crowd flow occurs since the ineluctable missing values. This problem is further complicated by the inherent scarcity and noise nature of OD crowd flow data. In this paper, we propose an uncertainty-aware interpolative and explainable framework, namely UApex, for realizing reliable and trustworthy OD crowd flow interpolation. Specifically, we first design a Variational Multi-modal Recurrent Graph Auto-Encoder (VMR-GAE) for uncertainty-aware OD crowd flow interpolation. A key idea here is to formulate the problem as semi-supervised learning on directed graphs. Next, to mitigate the data scarcity, we incorporate a distribution alignment mechanism that can introduce supplementary modals into variational inference. Then, a dedicated decoder with a Poisson prior is proposed for OD crowd flow interpolation. Moreover, to make VMR-GAE more trustworthy, we develop an efficient and uncertainty-aware explainer that can provide explanations from the spatiotemporal topology perspective via the Shapley value. Extensive experiments on two real-world datasets validate that VMR-GAE outperforms the state-of-the-art baselines. Also, an exploratory empirical study shows that the proposed explainer can generate meaningful spatiotemporal explanations.

In social service centers, practitioners engage in conversations with clients with dementia to facilitate their daily activities and provide support when they are distressed. However, the nature of the care demands the practitioner’s active engagement, which becomes difficult to deliver as the number of people who need care expands. Researchers have been investigating the efficacy of developing agents that assume conversational tasks to alleviate this work. To contribute to the future design of agents for caregiving, we collected and analyzed ten conversations between clients with mild dementia and practitioners who provide care. Our analyses of turn-taking dynamics and dialogue acts with 15k utterances uncovered patterns such as noticeable differences in clients’ and practitioners’ conversational dynamics and the prevalence of neutral-toned, question-oriented utterances by practitioners. We then prototyped a large language model-based script that generates responses to client utterances. We found potential approaches and challenges for making its utterance pattern more similar to that of a practitioner.

Path planning is a key enabling technology for drone-based applications, where the drone may encounter situations that require impromptu decisions to avoid task failure. Traditional works usually model it as an optimization problem with provable guarantees in a known map. However, they overlook the real-world uncertain dynamism (e.g. wind, illumination) stemming from the inadequate and inaccurate environmental information, which could pose risks to decision making and reduce the utility of path planning. To alleviate this uncertainty, we propose a Dynamic Environment-driven Path planning-oriented Deep Reinforcement Learning (DEP-DRL) paradigm that can be extended to more unfamiliar environments. We manage to do this by innovating in the following aspects: First, due to the influence of wind on the drone aerodynamic characteristics, we design a predictive drone state representation method which performs probabilistic wind-turbulence reasoning over drone kinematics in a deep reinforcement learning loop to conduct more realistic exploration. Then, we derive a system state transition mechanism for predicting future environmental observations and drone actions to minimize the potential system uncertainty and increase the path-planning success rate. We further propose an adaptive value updating strategy according to environmental illumination-based assessment in an online fashion. The experiments over different environments demonstrate the effectiveness of our method.

Vehicular Ad-hoc Networks (VANETs) constitute crucial elements within intelligent transportation systems. However, the rapid development of VANETs has brought forth an increasing number of security concerns. Conditional Privacy-Preserving Certificateless Signature (CPP-CLS) has emerged as a promising solution to ensure data security, preserve vehicle anonymity, and establish unlinkability in VANETs. In contrast to traditional public key infrastructure systems that involve cumbersome certificate management, and identity-based frameworks fraught with key escrow issues, CPP-CLS presents a more apt approach for VANETs. Unfortunately, the researches on CPP-CLS present a strange phenomenon in that a scheme proposed is always pointed out to have various security problems, especially public key replacement attacks. Moreover, there is a scarcity of published researches on the generic construction of CPP-CLS. To tackle these challenges, this paper proposes the first generic construction for CPP-CLS based on Type-T (Three-move type) signature, in which the public key reconstruction technique enables any receiver who owns a part of the sender’s public key and the KGC’s public key to reconstruct the complete sender’s public key, which can alleviate the public key replacement attacks. A formal security analysis proves that our scheme effectively guards against existential forgery under adaptively chosen message attacks in the random oracle model, contingent upon the security of the underlying Type-T signature. Furthermore, We provide two specific instantiations of the generic construction to verify feasibility. Among them, the instantiation based on module learning with errors is effective against quantum attacks. Based on extensive experimental results and theoretical analysis, our implementations surpass the majority of existing similar schemes in either performance or security. This substantiates the feasibility of our generic scheme, making it applicable for constructing CPP-CLS schemes.

Point-of-Interest (POI) recommendation, an important research hotspot in the field of urban computing, plays a crucial role in urban construction. While understanding the process of users’ travel decisions and exploring the causality of POI choosing is not easy due to the complex and diverse influencing factors in urban travel scenarios. Moreover, the spurious explanations caused by severe data sparsity, i.e., misrepresenting universal relevance as causality, may also hinder us from understanding users’ travel decisions. To this end, in this article, we propose a factor-level causal explanation generation framework based on counterfactual data augmentation for user travel decisions, named Factor-level Causal Explanation for User Travel Decisions (FCE-UTD), which can distinguish between true and false causal factors and generate true causal explanations. Specifically, we first assume that a user decision is composed of a set of several different factors. Then, by preserving the user decision structure with a joint counterfactual contrastive learning paradigm, we learn the representation of factors and detect the relevant factors. Next, we further identify true causal factors by constructing counterfactual decisions with a counterfactual representation generator, in particular, it can not only augment the dataset and mitigate the sparsity but also contribute to clarifying the causal factors from other false causal factors that may cause spurious explanations. Besides, a causal dependency learner is proposed to identify causal factors for each decision by learning causal dependency scores. Extensive experiments conducted on three real-world datasets demonstrate the superiority of our approach in terms of check-in rate, fidelity, and downstream tasks under different behavior scenarios. The extra case studies also demonstrate the ability of FCE-UTD to generate causal explanations in POI choosing.

Long-term urban crowd flow prediction involving the evolution trends of crowd flow is of great importance of traffic management, public safety and urban planning. However, learning long-term crowd flow is very challenging due to the latent effect of varied urban Point-of-Interests distribution, which is quite different from the short-term crowd flow mainly influenced by readily available external factors like weather, date, etc. The key issue for us is how to learn the interaction between POI distribution and human mobility in a dynamic way. To address this problem, we propose a POI-flow interaction based spatial-temporal framework (PFIST) for long-term crowd flow prediction. First, we model the long-term evolution representations of crowd flow and POI distribution. Then we study the dynamic interaction between POI transition patterns and crowd flow variation on different POI periods and categories. Afterwards, we decompose the flow sequence into long-term trend and daily variation parts and apply the normalized POI-flow interaction attention to the long-term trend parts. Finally, we model the spatial and multi-scale temporal dependencies to predict long-term crowd flow. Extensive experiments on Beijing map query track dataset and NYC taxi dataset demonstrate the superiority of PFIST.

Age of information (AoI) is an important metric used to quantify the freshness of data. By utilizing resources of the edge server near the source nodes, mobile edge computing (MEC) can speed up the processing of information updates and ensure data freshness. However, the resources of the edge server are usually limited, so it is necessary to study the resource allocation strategy to ensure data freshness and optimize the profit of the edge server. In this paper, we propose a game-theoretic approach for resource allocation in mobile edge computing to guarantee information freshness. First, with the purpose of ensuring information freshness, we formalize the problem as minimizing the computational cost of source nodes and maximizing the profit of the edge server. Then, we introduce a two-stage dynamic game model to simulate the competitive process. We further transform the resource allocation problem into a knapsack problem and propose an iterative resource allocation algorithm based on dynamic programming. Experimental results show that the proposed algorithm can obtain a Nash equilibrium and maximize the profit of the edge server while ensuring information freshness.

One of the most difficult data flow errors to detect caused by single-event upsets in space radiation is the Silent Data Corruption (SDC). To solve the problem of multi-bit upsets causing program SDC, an instruction multi-bit SDC vulnerability prediction model based on one-class support vector machine classification is built using SDC vulnerability analysis, which has more accurate vulnerability instruction identification capabilities. By hardening the program with selective instruction redundancy, we propose a multi-bit data flow error detection method for detecting SDC error (SDCVA-OCSVM), aiming to protect the data in the memory or register used by the program. We have also verified the effectiveness of the method through comparative experiments. The method has been verified to have a higher error detection rate and lower code size and time overhead.

As a promising learning paradigm for addressing the data scarcity and distribution mismatch issues, cross-domain prediction aims to leverage the transferable knowledge from the source domain to solve the learning problems in the target domain. Indeed, many urban computing tasks, such as cross- city/mode traffic flow prediction, have to face the severe data scarcity problem due to the heterogeneity in different data sources as well as the imbalanced development among cities. To this end, in this paper, we propose a cross-domain learning framework, namely CCMHC, which exploits Hierarchical Correlation between domains for Cross-City cross-Mode traffic flow prediction. Specifically, we first measure the correlation among inter-city traffic flows by exploring the similarity of region functions and road-networks. In this step, we filter out the regions with lower transfer ability from the source city to the target city. Then, we calculate the temporal correlations of traffic flows across different modes to select a source region that is highly related to the target region in a dynamic way. Moreover, a cross-domain urban flow prediction method is devised by transferring shared knowledge from the source city to the target city. Finally, experimental results on real-world data demonstrate the superiority of CCMHC over the state-of-the-art transfer learning methods. In addition, the generalization ability of the CCMHC framework on different neural network-based models is also validated.

Mobile onboard target detection system with autonomous drone-assisted Internet of Things, due to its inherent agility and coverage-effective deployment, is beneficial in city management, ecosystem monitoring, etc. However, most advanced detection methods become inefficient or even malfunction, since the computation resources for online detection in such a high-altitude dynamic environment are far beyond the capability of the drone. To this end, we design and implement ODTDS—an online drone-based target detection system that performs online data processing and autonomous navigation simultaneously with restricted resources. Specifically, to prolong detection durations with limited energy providing for continuous processing and flying, we develop an adaptive motion planner for autonomous and energy-efficient navigation. Meanwhile, to perform online target detection from complex environments, we propose a hybrid method integrating feature pyramid feedback with the speed up robust features, to enhance the background information to achieve high accuracy on the restricted computation platform of the drone. Based on these two designs, our system can address resource-constrained challenges to accomplish detection missions autonomously. Finally, we implement an ODTDS prototype and evaluate it through outdoor high-altitude experiments. The results of various performance evaluations can demonstrate the effectiveness of our system.

Nowadays, with the development of various kinds of sensors in smartphones or wearable devices, human activity recognition (HAR) has been widely researched and has numerous applications in healthcare, smart city, etc. Many techniques based on hand-crafted feature engineering or deep neural network have been proposed for sensor based HAR. However, these existing methods usually recognize activities offline, which means the whole data should be collected before training, occupying large-capacity storage space. Moreover, once the offline model training finished, the trained model can't recognize new activities unless retraining from the start, thus with a high cost of time and space. In this paper, we propose a multi-modality incremental learning model, called HarMI, with continuous learning ability. The proposed HarMI model can start training quickly with little storage space and easily learn new activities without storing previous training data. In detail, we first adopt attention mechanism to align heterogeneous sensor data with different frequencies. In addition, to overcome catastrophic forgetting in incremental learning, HarMI utilizes the elastic weight consolidation and canonical correlation analysis from a multi-modality perspective. Extensive experiments based on two public datasets demonstrate that HarMI can achieve a superior performance compared with several state-of-the-arts.

Unlike the traditional dock-based systems, dockless bike-sharing systems are more convenient for users in terms of flexibility. However, the flexibility of these dockless systems comes at the cost of management and operation complexity. Indeed, the imbalanced and dynamic use of bikes leads to mandatory rebalancing operations, which impose a critical need for effective bike traffic flow prediction. While efforts have been made in developing traffic flow prediction models, existing approaches lack interpretability, and thus have limited value in practical deployment. To this end, we propose an Interpretable Bike Flow Prediction (IBFP) framework, which can provide effective bike flow prediction with interpretable traffic patterns. Specifically, by dividing the urban area into regions according to flow density, we first model the spatio-temporal bike flows between regions with graph regularized sparse representation, where graph Laplacian is used as a smooth operator to preserve the commonalities of the periodic data structure. Then, we extract traffic patterns from bike flows using subspace clustering with sparse representation to construct interpretable base matrices. Moreover, the bike flows can be predicted with the interpretable base matrices and learned parameters. Finally, experimental results on real-world data show the advantages of the IBFP method for flow prediction in dockless bike sharing systems. In addition, the interpretability of our flow pattern exploitation is further illustrated through a case study where IBFP provides valuable insights into bike flow analysis.

Out-of-town recommendation is designed for those users who leave their home-town areas and visit the areas they have never been to before. It is challenging to recommend Point-of-Interests (POIs) for out-of-town users since the out-of-town check-in behavior is determined by not only the user’s home-town preference but also the user’s travel intention. Besides, the user’s travel intentions are complex and dynamic, which leads to big difficulties in understanding such intentions precisely. In this paper, we propose a TRAvel-INtention-aware Out-of-town Recommendation framework, named TRAINOR. The proposed TRAINOR framework distinguishes itself from existing out-of-town recommenders in three aspects. First, graph neural networks are explored to represent users’ home-town check-in preference and geographical constraints in out-of-town check-in behaviors. Second, a user-specific travel intention is formulated as an aggregation combining home-town preference and generic travel intention together, where the generic travel intention is regarded as a mixture of inherent intentions that can be learned by Neural Topic Model (NTM). Third, a non-linear mapping function, as well as a matrix factorization method, are employed to transfer users’ home-town preference and estimate out-of-town POI’s representation, respectively. Extensive experiments on real-world data sets validate the effectiveness of the TRAINOR framework. Moreover, the learned travel intention can deliver meaningful explanations for understanding a user’s travel purposes.

Potential crowd flow prediction for new planned transportation sites is a fundamental task for urban planners and administrators. Intuitively, the potential crowd flow of the new coming site can be implied by exploring the nearby sites. However, the transportation modes of nearby sites (e.g. bus stations, bicycle stations) might be different from the target site (e.g. subway station), which results in severe data scarcity issues. To this end, we propose a data-driven approach, named MOHER, to predict the potential crowd flow in a certain mode for a new planned site. Specifically, we first identify the neighbor regions of the target site by examining the geographical proximity as well as the urban function similarity. Then, to aggregate these heterogeneous relations, we devise a cross-mode relational GCN, a novel relation-specific transformation model, which can learn not only the correlation but also the differences between different transportation modes. Afterward, we design an aggregator for inductive potential flow representation. Finally, an LTSM module is used for sequential flow prediction. Extensive experiments on real-world data sets demonstrate the superiority of the MOHER framework compared with the state-of-the-art algorithms.

Higher-resolution biopsy slice images reveal many details, which are widely used in medical practice. However, taking high-resolution slice images is more costly than taking low-resolution ones. In this paper, we propose a joint framework containing a novel transfer learning strategy and a deep super-resolution framework to generate high-resolution slice images from low-resolution ones. The super-resolution framework called SRFBN+ is proposed by modifying a state-of-the-art framework SRFBN. Specifically, the structure of the feedback block of SRFBN was modified to be more flexible. Besides, it is challenging to use typical transfer learning strategies directly for the tasks on slice images, as the patterns on different types of biopsy slice images are varying. To this end, we propose a novel transfer learning strategy, called Channel Fusion Transfer Learning (CF-Trans). CF-Trans builds a middle domain by fusing the data manifolds of the source domain and the target domain, serving as a springboard for knowledge transfer. Thus, in the transfer learning setting, SRFBN+ can be trained on the source domain and then the middle domain and finally the target domain. Experiments on biopsy slice images validate SRFBN+ works well in generating super-resolution slice images, and CF-Trans is an efficient transfer learning strategy.

Various sensing and computing technologies have gradually outlined the future of the intelligent city. Passenger flow prediction of public transports has become an important task in Intelligent Transportation System (ITS), which is the prerequisite for traffic management and urban planning. There exist many methods based on deep learning for learning the spatiotemporal features from high non-linearity and complexity of traffic flows. However, they only utilize temporal correlation and static spatial correlation, such as geographical distance, which is insufficient in the mining of dynamic spatial correlation. In this paper, we propose the Dynamic-Static-based Spatiotemporal Multi-Graph Neural Networks model (DSSTMG) for predicting traffic passenger flows, which can concurrently incorporate the temporal and multiple static and dynamic spatial correlations. Firstly, we exploit the multiple static spatial correlations by multi-graph fusion convolution operator, including adjacent relation, station functional zone similarity and geographical distance. Secondly, we exploit the spatial dynamic correlations by calculating the similarity between the flow pattern of stations over a period of time, and build the dynamic spatial attention. Moreover, we use time attention and encoder-decoder architecture to capture temporal correlation. The experimental results on two realworld datasets show that the proposed DSSTMG outperforms state-of-the-art methods.

While Point-of-Interest (POI) recommendation has been a popular topic of study for some time, little progress has been made for understanding why and how people make their decisions for the selection of POIs. To this end, in this paper, we propose a user decision profiling framework, named PROUD, which can identify the key factors in people's decisions on choosing POIs. Specifically, we treat each user decision as a set of factors and provide a method for learning factor embeddings. A unique perspective of our approach is to identify key factors, while preserving decision structures seamlessly, via a novel scalar projection maximization objective. Exactly solving the objective is non-trivial due to a sparsity constraint. To address this, our PROUD adopts a self projection attention and an L2 regularized sparse activation to directly estimate the likelihood of each factor to be a key factor. Finally, extensive experiments on real-world data validate the advantage of PROUD in preserving user decision structures. Also, our case study indicates that the identified key decision factors can help us to provide more interpretable recommendations and analyses.

The Aerial Internet-of-Things (Aerial-IoT) systems, deploying sensors on high-altitude platforms, e.g., drones, parachutes, and aircrafts, are a crucial monitor due to its agile maneuverability and augmentation of observation, collection, and communication. As such, the measurement accuracy and requirements of Aerial-IoT are far beyond the ability of general commercial-off-the-shelf sensors, especially in the high-altitude environment, where environmental factors (air pressure, temperature, humidity, wind movement, etc.) tend to change rapidly and lead to highly deviated readings. In this article, we tackle this challenge. First, we introduce our designed measurement system for Aerial-IoT. Then, to compensate for the low data quality and calibrate the deviation data from sensors, we take into account the inherent correlations and interaction between sensor data and environmental factors, and construct a data calibration model, called data calibration based on the neural network (DC-NN). Finally, to illustrate the effectiveness of our system, we carry out a real-world implementation by deploying sensors on the surface of parachutes in a dynamic airdrop environment. Extensive experiments on temperature-humidity-material-tensile-testing (THMTT) and high-altitude airdrop are conducted to show the significant improvements of our proposed DC-NN model.

Multi-modal transportation recommendation has a goal of recommending a travel plan which considers various transportation modes, such as walking, cycling, automobile, and public transit, and how to connect among these modes. The successful development of multi-modal transportation recommendation systems can help to satisfy the diversified needs of travelers and improve the efficiency of transport networks. However, existing transport recommender systems mainly focus on unimodal transport planning. To this end, in this paper, we propose a joint representation learning framework for multi-modal transportation recommendation based on a carefully-constructed multi-modal transportation graph. Specifically, we first extract a multi-modal transportation graph from large-scale map query data to describe the concurrency of users, Origin-Destination (OD) pairs, and transport modes. Then, we provide effective solutions for the optimization problem and develop an anchor embedding for transport modes to initialize the embeddings of transport modes. Moreover, we infer user relevance and OD pair relevance, and incorporate them to regularize the representation learning. Finally, we exploit the learned representations for online multimodal transportation recommendations. Indeed, our method has been deployed into one of the largest navigation Apps to serve hundreds of millions of users, and extensive experimental results with real-world map query data demonstrate the enhanced performance of the proposed method for multimodal transportation recommendations.

Human mobility data accumulated from Point-of-Interest (POI) check-ins provides great opportunity for user behavior understanding. However, data quality issues (e.g., geolocation information missing, unreal check-ins, data sparsity) in real-life mobility data limit the effectiveness of existing POIoriented studies, e.g., POI recommendation and location prediction, when applied to real applications. To this end, in this paper, we develop a model, named Bi-STDDP, which can integrate bi-directional spatio-temporal dependence and users’ dynamic preferences, to identify the missing POI check-in where a user has visited at a specific time. Specifically, we first utilize bi-directional global spatial and local temporal information of POIs to capture the complex dependence relationships. Then, target temporal pattern in combination with user and POI information are fed into a multi-layer network to capture users’ dynamic preferences. Moreover, the dynamic preferences are transformed into the same space as the dependence relationships to form the final model. Finally, the proposed model is evaluated on three large-scale real-world datasets and the results demonstrate significant improvements of our model compared with state-of-the-art methods. Also, it is worth noting that the proposed model can be naturally extended to address POI recommendation and location prediction tasks with competitive performances.

Nowadays the system vulnerability caused by soft errors grows exponentially, of which Silent Data Corruption(SDC) is one of the most harmful issues due to introducing unnoticed changes to the original data and error outputs. Thus, the detection of SDC-causing errors is extremely significant to the system reliability. However, most of the current detecting techniques require sufficient data of fault injections for training, which are difficult to achieve in practice because of high resources consumption, such as expensive execution time and code size costs. To this end, we propose a lightweight model named Deep Forest Regression based Multi-granularity Redundancy(DFRMR) to improve the error detection rate and meanwhile decrease the resources consumption. Specifically, first, we employ the program analysis to extract instruction features which are highly related to SDCs. Second, we design the deep forest regression model to predict the SDC vulnerability of instructions. Third, we optimize the error detection procedure by duplicating the critical instructions with different granularity. Finally, we evaluate our DFRMR model on Mibench benchmarks with multiple testing programs. The results show that our method attains better detection accuracy compared to other state-of-the-art methods and keeps the low multi-granularity redundancy.

Autonomous navigation systems on unmanned aerial vehicles (UAVs) equipped with multiple sensors are essential to various applications in the smart city and intelligent transportation. However, the general autonomous navigation models are markedly influenced by the prior knowledge from training environments, which in turn are not applicable in unknown environments. To address this issue, we propose an online autonomous UAV navigation system named as multi-sensor data-fusion-based autonomous navigation (MDFAN) system for unknown flight environments, including the collision avoidance and path planning. Specifically, first, the newly MDFAN system formulates the navigation problem as a decision-making path planning problem to reduce the dependence of prior knowledge of the flight environment. Secondly, we develop a multi-sensor data-fusion-based method to extract more effective local environment information for mining the inherent inter-relationship between the local environment information and the current state of the UAV. Thirdly, we propose a deep reinforcement learning method for handling uncertain situations of the unknown environment. Finally, we validated our method both on the simulated and real-world environments.

Detection of moving targets is a basic but challenging function of drone-based surveillance systems (DBSSs), which could give rise to various potential applications in smart city and intelligent transportation. However, how to detect moving targets in the dynamic high-altitude environment with restricted computational resources is one of the most critical ones. Furthermore, during the detection, the moving target could travel in dynamic speed and be blocked by dense-obstructer, such as woods and buildings. In this paper, we develop an online drone-based moving target detection (ODMTD)system, which performs moving target detection in dense-obstructer areas. Specifically, first, our proposed system simultaneously performs adaptive path planning and autonomous drone flight via the combination of historical path cost and energy loss computation by using drone attitudes. Second, to detect moving targets in the dynamic background, we develop an algorithm of combing the speed up robust features and approximate nearest neighbors, shortly SURF-ANN, for estimating and compensating the global motion of the background. Finally, in order to calibrate distorted images taken by the camera obliquely, we utilize perspective transformation to remap images into another plane, and then detect moving targets by subtracting registered images (SRI). Furthermore, real-time outdoor high-altitude experiments, by comprising with the state-of-art methods, demonstrate the effectiveness of our ODMTD system.

Urban Region-of-Interest (ROI) refers to the integrated urban areas with specific functionalities that attract people's attentions and activities, such as the recreational business districts, transportation hubs, and city landmarks. Indeed, at the macro level, ROI is one of the representatives for agglomeration economies, and plays an important role in urban business planning. At the micro level, ROI provides a useful venue for understanding the urban lives, demands and mobilities of people. However, due to the vague and diversified nature of ROI, it still lacks of quantitative ways to investigate ROIs in a holistic manner. To this end, in this paper we propose a systematic study on ROI analysis through mining the large-scale online map query logs, which provides a new data-driven research paradigm for ROI detection and profiling. Specifically, we first divide the urban area into small region grids, and calculate their PageRank value as visiting popularity based on the transition information extracted from map queries. Then, we propose a density-based clustering method for merging neighboring region grids with high popularity into integrated ROIs. After that, to further explore the profiles of different ROIs, we develop a spatial-temporal latent factor model URPTM (Urban Roi Profiling Topic Model) to identify the latent travel patterns and Point-of-Interest (POI) demands of ROI visitors. Finally, we implement extensive experiments to empirically evaluate our approaches based on the large-scale real-world data collected from Beijing. Indeed, by visualizing the results obtained from URPTM, we can successfully obtain many meaningful travel patterns and interesting discoveries on urban lives.

ABSTRACT Signal‐strength‐based location estimation in wireless sensor networks is to locate the physical positions of unknown sensorsviathe received signal strengths. In this field, there are few localization researches sufficiently exploiting topology structures of the network in both signal space and physical space. The goal of this paper is to first establish two effective localization models based on specific manifold (or local) structures of both signal space and physical (location) space by using our previous locality preserving canonical correlation analysis (LPCCA) model and a newly‐proposed locality correlation analysis (LCA) model, and then develop their corresponding novel location algorithms, called location estimation—LPCCA (LE—LPCCA) and location estimation—LCA (LE—LCA). Since both LPCCA and LCA relatively sufficiently take into account locality characteristics of the manifold structures in both the spaces, our localization algorithms developed from them consequently achieve better localization accuracy than other publicly available advanced algorithms. Copyright © 2011 John Wiley & Sons, Ltd.

Localizing sensors based on Received Signal Strength Indicator (RSSI) localization technique in wireless sensor network can be treated as building a mapping between signal and physical spaces, and the mapping is established from a set of given paired signal strengths and physical location data of known sensors. However, in some realistic scenarios, such a set of completely-paired sensor data is not always accessible, which brings a big challenge for localization of sensors. The localization research in such a scenario is currently almost ignored. In this paper, we develop a novel algorithm to tackle this problem in localization with paired as well as many unpaired data by adapting our previously-proposed Locality Correlation Analysis model; the new algorithm is named as Partially Paired Locality Correlation Analysis (PPLCA). Experimental results in both outdoor and indoor environments do show the feasibility and effectiveness of the proposed algorithm.