Humanoid robots are promising to learn a diverse set of human-like locomotion behaviors, including standing up, walking, running, and jumping. However, existing methods predominantly require training independent policies for each skill, yielding behavior-specific controllers that exhibit limited generalization and brittle performance when deployed on irregular terrains and in diverse situations. To address this challenge, we propose Adaptive Humanoid Control (AHC) that adopts a two-stage framework to learn an adaptive humanoid locomotion controller across different skills and terrains. Specifically, we first train several primary locomotion policies and perform a multi-behavior distillation process to obtain a basic multi-behavior controller, facilitating adaptive behavior switching based on the environment. Then, we perform reinforced fine-tuning by collecting online feedback in performing adaptive behaviors on more diverse terrains, enhancing terrain adaptability for the adaptive behavior controller. We conduct experiments in both simulation and real-world experiments in Unitree G1 robots. The results show that our method exhibits strong adaptability across various situations and terrains.

Intuitively, an ideal collaborative filtering (CF) model should learn from users' full rankings over all items to make optimal top-K recommendations. Due to the absence of such full rankings in practice, most CF models rely on pairwise loss functions to approximate full rankings, resulting in an immense performance gap. In this paper, we provide a novel analysis using the multiple ordinal classification concept to reveal the inevitable gap between a pairwise approximation and the ideal case. However, bridging the gap in practice encounters two formidable challenges: (1) none of the real-world datasets contains full ranking information; (2) there does not exist a loss function that is capable of consuming ranking information. To overcome these challenges, we propose a pseudo-ranking paradigm (PRP) that addresses the lack of ranking information by introducing pseudo-rankings supervised by an original noise injection mechanism. Additionally, we put forward a new ranking loss function designed to handle ranking information effectively. To ensure our method's robustness against potential inaccuracies in pseudo-rankings, we equip the ranking loss function with a gradient-based confidence mechanism to detect and mitigate abnormal gradients. Extensive experiments on four real-world datasets demonstrate that PRP significantly outperforms state-of-the-art methods.

The cold-start problem remains a long-standing challenge in recommender systems. Recent advances in large language models (LLMs) have opened new avenues for addressing cold-start scenarios through data augmentation. However, existing cold-start augmentation methods often suffer from negative augmentation, manifesting as incomplete augmentation, where generated interactions fail to comprehensively reflect user preferences, and inaccurate augmentation, where they conflict with user intent. These issues largely stem from two limitations: (1) the inability to effectively incorporate collaborative signals, which are critical for preference alignment, and (2) the lack of awareness of the downstream model's learning dynamics during data augmentation. To the best of our knowledge, the latter has not been studied in the literature.

In real-world scenarios, extreme cases where pedestrians suddenly emerge from blind spots or occlusions, leaving only a minimal amount of observable trajectory points, occur frequently. This presents a significant challenge for autonomous driving and robotic navigation, where pedestrian safety and timely response are critical considerations. To address this challenge, we propose a framework for instantaneous trajectory prediction using Latent bidirectional Cooperative Diffusion (LCD). It designs a complementary mechanism that constructs a coupled bidirectional cooperative diffusion model. LCD simultaneously and progressively generates unobserved past trajectories and future trajectories, feeding each other as conditions into the cross-attention module for mutual guidance. This framework employs CVAE as its encoder to map the observed multi-model trajectories into a high-dimensional latent space to enhance complex representations. Experiments conducted on the ETH/UCY and SDD datasets demonstrate the superiority of our framework.

Modeling stochastic multi-ship trajectories is vital for maritime safety and interaction efficiency. Recent researches show that diffusion models excel in trajectory prediction, surpassing GANs and VAEs in generation quality, diversity and stability. However, their slow sampling speed remains a major limitation, as producing high-quality trajectories typically requires hundreds of denoising steps. We introduce DDA, a novel method that accelerates multi-ship trajectory generation by distilling the reverse diffusion process, progressively reducing sampling steps by half while minimizing quality loss. We use CVAE-based encoder to map multimodal inputs into state embeddings in the latent space, and use distillation diffusion in the latent space to more quickly and better represent multi-ship trajectories. The diffusion model uses Transformer-based core, and we incorporate SO(2) invariance and equivariance to enhance model representation. Validation on real-world AIS datasets shows that the student model retains high-quality trajectory generation while sampling speed is approximately 30 times faster.

In the field of big data, the phenomenon of missing data is widespread, particularly pronounced in time-series datasets, where its impact on subsequent processing is significant. However, existing data imputation methods have not comprehensively and adequately addressed the bidirectional spatio-temporal dependencies within the data, especially the dependencies among different dimensions at various timestamps in multivariate time-series data. To address this issue, this paper proposes a hybrid model that integrates Graph Neural Networks (GNNs) and Bidirectional Recurrent Neural Networks (RNNs) for imputing missing values in multivariate time-series data. The model introduces an enhanced bidirectional Gated Recurrent Unit (GRU) to capture the bidirectional temporal dependencies in the data. Additionally, it employs graph structure learning and attention mechanisms to construct a correlation graph, aggregating inter-dimensional correlation information through graph neural networks. Experimental results indicate that our model has significantly improved the imputation accuracy on three public datasets, particularly under conditions of low missing rates.

Entity Resolution (ER), a crucial task in data integration and cleaning, identifies records from different sources that refer to the same entity. Despite advances in existing state-of-the-art (SOTA) methods, a significant challenge remains largely unaddressed: the quality of the data used for training these models. This paper shifts the focus from model-centric improvements to data-centric enhancements, identifying two primary data quality issues: (1) low-hardness positive samples and (2) the scarcity of positive samples relative to negative samples. To tackle these challenges, we propose a novel Data Quality Enhancement Framework (DQEF) comprising an adversarial hardness enhancement method and a noise perturbation-based positive sample generation method. DQEF generates informative and hard positive samples, improving the model’s training effectiveness. As a model-agnostic approach, it can seamlessly integrate with mainstream ER models to enhance performance. Experiments on multiple ER models and real-world datasets show that DQEF significantly boosts performance, surpassing current SOTA results.

Accurate spatiotemporal prediction is crucial for improving public health and enhancing travel efficiency. However, this task is challenging due to the complex spatiotemporal dependencies between regions. Most methods learn spatial representations without considering spatially skewed distributions, where data from the different regions often exhibit long-tailed distributions, limiting the spatial representation capabilities of the tail distribution regions. To address the above challenges, we propose a novel Multi-level Cross-view Graph Convolution Network (MCG-STNet) for spatiotemporal prediction. The core advantage of MCG-STNet is its ability to jointly capture non-Euclidean dependencies from global and local views while enhancing the spatial feature representation of tail distribution regions through an adaptive cross-view information transfer network. Specifically, we design a fuzzy neural network-based soft clustering approach to capture the functional relationships of regions with long-tail distributed data in the global view, integrating domain expertise to capture spatial dependencies among regional functionalities representations adaptively. We propose an adaptive cross-view information transfer network, allowing for the adaptive extraction of supplementary spatial feature representations from global to local view. Experimental results on five real-world datasets demonstrate that our proposed MCG-STNet outperforms other baselines. Further analysis validates the insights into the effectiveness of MCG-STNet.

Contrastive learning (CL) has achieved remarkable success in addressing data sparsity issues in collaborative filtering (CF) for recommender systems (RSs). The key principle is to generate different augmented views given a user-item interaction graph. However, prior endeavors mainly focus on performing augmentation via stochastic functions, e.g., by injecting perturbations into different hidden dimensions uniformly. Without fine control, the hidden representations of augmentations may contain noisy dimensions that are harmful to CL and irrelevant to RSs. Removing dimension-specific noise is a challenging task due to the following two major bottlenecks. It is difficult to (i) distinguish different dimensions' efficacy for CL and (ii) bridge the semantic gap between CL and RSs. Overlooking these limitations may cause redundant, false-positive, and irrelevant noise in hidden dimensions of the augmented views.

Optimizer-based meta-learning, specifically model-agnostic meta-learning (MAML), has emerged as a powerful tool for tackling the cold-start recommendation problem. In these meta-learning-based methods, recommendations for individual users are typically treated as separate tasks and learned independently. However, this task-by-task learning paradigm presents several observable limitations. First, learning one task at a time ignores inter-task correlations, i.e., collaborative signals, which limits the meta-model's receptive field and prevents it from leveraging valuable shared information, ultimately leading to subpar performance. Second, the meta-model is susceptible to the task distribution, i.e., the varied preference distributions among different users, which in turn introduces biases and inconsistencies, resulting in a less robust model that may perform well on certain user groups while underperforming on others. In this paper, we explore the correlations among different tasks in cold-start recommendations and develop a novel strategy termed cross-task collaborative meta-learning (CCML). More specifically, we propose a collaborative task sampling module designed to mitigate the adverse impact of irrelevant tasks during meta-model learning. This module adaptively identifies tasks that are both similar and beneficial to the primary task, ensuring that the meta-model learns from relevant and supportive information. Additionally, to harness collaborative information across relevant tasks, we introduce a bi-level cross-task meta-training strategy. This strategy leverages multi-task learning to capture collaborative knowledge simultaneously and enhance user profiling with pertinent information. Extensive experiments on four public benchmark datasets demonstrate the advantages of CCML over many state-of-the-art cold-start recommendation methods. Our results show significant improvements in recommendation accuracy and robustness, highlighting the potential of cross-task collaboration in enhancing meta-learning-based recommender systems. The code is available athttps://anonymous.4open.science/r/CCML-F064.

Negative sampling plays a crucial role in implicit-feedback-based collaborative filtering, where it leverages massive unlabeled data to generate negative signals for guiding supervised learning. The current state-of-the-art approaches focus on utilizing hard negative samples that contain more information to establish a better decision boundary. To strike a balance between efficiency and effectiveness, most existing methods adopt a two-pass approach: In the first pass, a fixed number of unobserved items are sampled using a simple static distribution, while, in the second pass, a more sophisticated negative sampling strategy is employed to select the final negative items. However, selecting negative samples solely from the original items in a dataset is inherently restricted due to the limited available choices and thus may not be able to effectively contrast positive samples. In this article, we empirically validate this observation through meticulously designed experiments and identify three major limitations of existing solutions: ambiguous trap, information discrimination, and false-negative samples. Our response to such limitations is to introduce “denoised” and “augmented” negative samples that may not exist in the original dataset. This direction renders a few substantial technical challenges. First, constructing augmented negative samples may introduce excessive noise that eventually distorts the decision boundary. Second, the scarcity of supervision signals hampers the denoising process. To this end, we introduce, to the best of our knowledge, a novel generic denoising and augmented negative sampling paradigm and provide a concrete instantiation. First, we disentangle the hard and easy factors of negative items. Then, we regulate the augmentation of easy factors by carefully considering the direction and magnitude. Next, we propose a reverse attention mechanism to learn a user’s negative preference, which allows us to perform a dimension-level denoising procedure on hard factors. Finally, we design an advanced negative sampling strategy to identify the final negative samples, taking into account both the score function used in existing methods and, to the best of our knowledge, a novel metric called synthesization gain. Through extensive experiments on real-world datasets, we demonstrate that our method substantially outperforms state-of-the-art baselines. Our code is publicly available at https://github.com/Asa9aoTK/ANS-Recbole .

Negative sampling is essential for implicit collaborative filtering to provide proper negative training signals so as to achieve desirable performance. We experimentally unveil a common limitation of all existing negative sampling methods that they can only select negative samples of a fixed hardness level, leading to the false positive problem (FPP) and false negative problem (FNP). We then propose a new paradigm called adaptive hardness negative sampling (AHNS) and discuss its three key criteria. By adaptively selecting negative samples with appropriate hardnesses during the training process, AHNS can well mitigate the impacts of FPP and FNP. Next, we present a concrete instantiation of AHNS called AHNS_{p

Numerous pervasive applications, such as guiding outdoor health activities and allocating public transportation resources, rely on accurate predictions. Intuitively, the readings of a region may be influenced by various relational patterns with other regions, which can be revealed by exploring the complex spatial correlations between them. Recent studies have demonstrated the potential for improving spatiotemporal prediction performance by modeling various spatial correlations as graphs and capturing spatial dependencies using parallel multi-graph convolution network methods. However, these methods cannot directly learn dynamic spatial dependencies across graphs, leading to insufficient utilization of spatial contextual information. To address the above issues, we propose a novel dynamic multi-graph convolution networks, named DyMGCN, for spatiotemporal forecasting. Specifically, we introduce fuzzy theory, which combines fuzzy systems and neural networks, to learn uncertain dynamic dependencies across graphs and represent them as weight matrices. We further improve the understanding and utilization of spatial correlations by capturing stable long-term autocorrelations between regions. Many experiments on multiple real-world spatiotemporal datasets demonstrate that our proposed DyMGCN outperforms other baselines in spatiotemporal prediction methods.

Traditional sequential recommendation methods assume that users' sequence data is clean enough to learn accurate sequence representations to reflect user preferences. In practice, users' sequences inevitably contain noise (e.g., accidental interactions), leading to incorrect reflections of user preferences. Consequently, some pioneer studies have explored modeling sequentiality and correlations in sequences to implicitly or explicitly reduce noise's influence. However, relying on only available intra-sequence information (i.e., sequentiality and correlations in a sequence) is insufficient and may result in over-denoising and under-denoising problems (OUPs), especially for short sequences. To improve reliability, we propose to augment sequences by inserting items before denoising. However, due to the data sparsity issue and computational costs, it is challenging to select proper items from the entire item universe to insert into proper positions in a target sequence. Motivated by the above observation, we propose a novel framework-Self-augmented Sequence Denoising for sequential Recommendation (SSDRec) with a three-stage learning paradigm to solve the above challenges. In the first stage, we empower SSDRec by a global relation encoder to learn multi-faceted inter-sequence relations in a data-driven manner. These relations serve as prior knowledge to guide subsequent stages. In the second stage, we devise a self-augmentation module to augment sequences to alleviate OUPs. Finally, we employ a hierarchical denoising module in the third stage to reduce the risk of false augmentations and pinpoint all noise in raw sequences. Extensive experiments on five real-world datasets demonstrate the superiority of SSDRec over state-of-the-art denoising methods and its flexible applications to mainstream sequential recommendation models. The source code is available online at https://github.com/zc-97/SSDRec.

Accurate ship trajectory prediction is crucial to ensure the safety and interaction efficiency of maritime traffic, especially in areas with dense ships and areas with intersecting waterways. Different from the certainty of short-term trajectory prediction, for long-term trajectory prediction, the uncertainty of direction and speed changes in operation is amplified, and the prediction of ship trajectory is more difficult. To solve this problem, we propose A framework based on a diffusion probability model balancing diversity and determinacy (DM-BDD). Inspired by the denoising process in the diffusion model, we gradually reduce the range of trajectories that the ships are navigable, and finally achieve a desired and deterministic trajectory. This process is learned using a parameterized Markov chain based on the currently observed trajectory. The length of the chain is adjusted to control the degree of uncertainty of the running trajectory and the final position, and to balance the diversity and certainty of the prediction results. We design an architecture consisting of a temporal-social encoding network and an LSTM-based decoder network to achieve the purpose of capturing the spatial dependence and social dependence of trajectories. We test and compare on the AIS dataset of the real Weihai Port and the Nantong waterway of the Yangtze River estuary. The experimental results show that the short, medium and long-term prediction results of the proposed model on the two data sets achieve the best performance in the field of ship trajectory prediction.

Collaborative filtering (CF) stands as a cornerstone in recommender systems, yet effectively leveraging the massive unlabeled data presents a significant challenge. Current research focuses on addressing the challenge of unlabeled data by extracting a subset that closely approximates negative samples. Regrettably, the remaining data are overlooked, failing to fully integrate this valuable information into the construction of user preferences. To address this gap, we introduce a novel positive-neutral-negative (PNN) learning paradigm. PNN introduces a neutral class, encompassing intricate items that are challenging to categorize directly as positive or negative samples. By training a model based on this triple-wise partial ranking, PNN offers a promising solution to learning complex user preferences. Through theoretical analysis, we connect PNN to one-way partial AUC (OPAUC) to validate its efficacy. Implementing the PNN paradigm is, however, technically challenging because: (1) it is difficult to classify unlabeled data into neutral or negative in the absence of supervised signals; (2) there does not exist any loss function that can handle set-level triple-wise ranking relationships. To address these challenges, we propose a semi-supervised learning method coupled with a user-aware attention model for knowledge acquisition and classification refinement. Additionally, a novel loss function with a two-step centroid ranking approach enables handling set-level rankings. Extensive experiments on four real-world datasets demonstrate that, when combined with PNN, a wide range of representative CF models can consistently and significantly boost their performance. Even with a simple matrix factorization, PNN can achieve comparable performance to sophisticated graph neutral networks. Our code is publicly available at https://github.com/Asa9aoTK/PNN-RecBole.

Abstract SpMV (Sparse matrix‐vector multiplication) is an important computing core in traditional high‐performance computing and also one of the emerging data‐intensive applications. For diagonal sparse matrices, it is frequently necessary to fill in a large number of zeros to maintain the diagonal structure as for using DIA (Diagonal) storage format. The fact that filling with zeros may consume additional computing and memory resources, will certainly lead to degradation of the parallel computing performance of SpMV, further causing computing and storage redundancy. To solve the deficiencies of the DIA format, a Two‐stage parallel SpMV method is presented in this paper, which can reasonably distribute the data of diagonal matrix and irregular matrix to different CUDA kernels. As different corresponding compression methods are particularly designed for different matrix forms, a partition‐based hybrid format of DIA and CSR (HPDC) is therefore adopted in the two‐stage method to ensure load balancing among computing resources and continuity of data access on the diagonal. Simultaneously, a standard deviation among blocks is used as a criterion to obtain the optimal number of blocks and distribution of data. The experimental data were implemented in the Florida data set. Compared to DIA, cuSPARSE‐CSR, HDC, and BRCSD, the execution time of the Two‐stage method is shortened by 4, 3.4, 1.9, and 1.15, respectively.

With the development of the Industrial Internet of Things (IoT), the work of large-scale data collection makes spatiotemporal crowdsensing (SC) play an important role. Mobile devices equipped with sensors could act as workers to collect and process data for uploading. In the task allocation process, a fully static allocation fails to meet the needs of realistic conditions, while a completely dynamic allocation fails to achieve the desired results. Therefore, we assume a task-scheduled execution scenario that combines the above two conditions. In the pre-allocation process, an original time location constraints (ORTA) allocation algorithm is first proposed. Then it is optimized (OPTA) to fully utilize the remaining time of the workers and increase the matched number. In addition, the design of the incentive mechanism is an effective means to improve the task completion rate of the platform. To efficiently utilize the limited platform budget in the long run, a Q-learning-based algorithm is proposed to identify target inspire tasks and subsequently increase their reward to attract workers’ participation. Finally, comparison experiments are conducted on real datasets to verify the effectiveness of our algorithm. Furthermore, the experiments on a Raspberry Pi local terminal are conducted under a satellite-based environment.

Sequential recommender systems have demonstrated a huge success for next-item recommendation by explicitly exploiting the temporal order of users' historical interactions. In practice, user interactions contain more useful temporal information beyond order, as shown by some pioneering studies. In this paper, we systematically investigate various temporal information for sequential recommendation and identify three types of advantageous temporal patterns beyond order, including absolute time information, relative item time intervals and relative recommendation time intervals. We are the first to explore item-oriented absolute time patterns. While existing models consider only one or two of these three patterns, we propose a novel holistic temporal pattern based neural network, named HTP, to fully leverage all these three patterns. In particular, we introduce novel components to address the subtle correlations between relative item time intervals and relative recommendation time intervals, which render a major technical challenge. Extensive experiments on three real-world benchmark datasets show that our HTP model consistently and substantially outperforms many state-of-the-art models. Our code is publically available at https://github.com/623851394/HTP/tree/main/HTP-main.

Machine learning algorithms are currently used in various fields. The success of these algorithms is following the access of large amounts of sensitive personal data, which raises sociological concerns about privacy and fairness. At present, existing works only focus on providing fairness guarantee for the specific attribute that “can affect fairness” without considering the impact of other attributes on that attribute. Therefore, this paper takes the classical logistic regression model as the research object, uses Decision Tree Analysis and Bayesian Network to complete the analysis, and uses functional mechanism to achieve fair privacy protection. For the functional mechanism, we analyze that the Chebyshev polynomial expansion is a less error approach. Theoretical analysis and experimental results show that proposed algorithms can effectively achieve differential privacy and fairness guarantee while maintaining good utility.

Cross-domain sequential recommendation (CDSR) aims to predict the next item that a user is most likely to interact with based on past sequential behavior from multiple domains. Existing works on CDSR usually transfer knowledge across different domains by linking items between domains via common users, which suffers from the following limitations: (1) due to the inherent differences between the domains, transferring information across domains can be affected by different representations of related items in different domains. (2) None of existing studies consider the time interval information among items, which is essential in sequential recommendation to capture user intents over time. In this work, we propose a novel cross-domain sequential recommendation model to address the above challenges. Specifically, we first design a shared proxy item encoder to generate a universal representation for each item in all domains by using its textual descriptions. Then, we develop a time-interval-aware attention encoder to represent sequences by considering the time interval information. Moreover, we present a contrastive learning auxiliary task to enhance a cross-domain sequence by weighing the importance of the items in the auxiliary domain with respect to the objective domain. Experiments demonstrate the superiority of our proposed method from various aspects.

Federated learning is a special kind of distributed machine learning, in which multiple clients work together to solve a machine learning problem with the collaboration of a central server, and the clients only need to upload parameters for server aggregation instead of uploading raw data, so the privacy of the clients can be protected. However, existing research shows that an attacker who obtains the parameters uploaded by the client can reverse the privacy information of the client, and Federated Learning applies a local differential privacy approach to protect the information of the parameters uploaded by the client from being leaked. However, this privacy protection approach assumes the same level of privacy protection for all clients. To the best of our knowledge, there needs to be work that satisfies the personalized privacy needs of clients. To address this problem, we propose a personalized local differential privacy-based federation framework that satisfies the personalized privacy needs of clients and better protects the privacy of clients by making the specific privacy needs of clients inaccessible to the server. We have conducted extensive experiments on six benchmark datasets, and our approach works better and achieves personalized privacy protection compared to the same privacy-preserving method.

Negative sampling is essential for implicit-feedback-based collaborative filtering, which is used to constitute negative signals from massive unlabeled data to guide supervised learning. The state-of-the-art idea is to utilize hard negative samples that carry more useful information to form a better decision boundary. To balance efficiency and effectiveness, the vast majority of existing methods follow the two-pass approach, in which the first pass samples a fixed number of unobserved items by a simple static distribution and then the second pass selects the final negative items using a more sophisticated negative sampling strategy. However, selecting negative samples from the original items in a dataset is inherently restricted due to the limited available choices, and thus may not be able to contrast positive samples well. In this paper, we confirm this observation via carefully designed experiments and introduce two major limitations of existing solutions: ambiguous trap and information discrimination.

Negative sampling is essential for implicit collaborative filtering to generate negative samples from massive unlabeled data. Unlike existing strategies that consider items as a whole when selecting negative items, we argue that normally user interactions are mainly driven by some relevant, but not all, factors of items, leading to a new direction of negative sampling. In this paper, we introduce a novel disentangled negative sampling (DENS) method. We first disentangle the relevant and irrelevant factors of positive and negative items using a hierarchical gating module. Next, we design a factor-aware sampling strategy to identify the best negative samples by contrasting the relevant factors while keeping irrelevant factors similar. To ensure the credibility of the disentanglement, we propose to adopt contrastive learning and introduce four pairwise contrastive tasks, which enable to learn better disentangled representations of the relevant and irrelevant factors and remove the dependency on ground truth. Extensive experiments on five real-world datasets demonstrate the superiority of DENS against several state-of-the-art competitors, achieving over 7% improvement over the strongest baseline in terms of [email protected] and [email protected] Our code is publically available at https://github.com/Riwei-HEU/DENS .

In practical recommendation scenarios, users often interact with items under multi-typed behaviors (e.g., click, add-to-cart, and purchase). Traditional collaborative filtering techniques typically assume that users only have a single type of behavior with items, making it insufficient to utilize complex collaborative signals to learn informative representations and infer actual user preferences. Consequently, some pioneer studies explore modeling multi-behavior heterogeneity to learn better representations and boost the performance of recommendations for a target behavior. However, a large number of auxiliary behaviors (i.e., click and add-to-cart) could introduce irrelevant information to recommenders, which could mislead the target behavior (i.e., purchase) recommendation, rendering two critical challenges: (i) denoising auxiliary behaviors and (ii) bridging the semantic gap between auxiliary and target behaviors. Motivated by the above observation, we propose a novel framework–Denoising and Prompt-Tuning (DPT) with a three-stage learning paradigm to solve the aforementioned challenges. In particular, DPT is equipped with a pattern-enhanced graph encoder in the first stage to learn complex patterns as prior knowledge in a data-driven manner to guide learning informative representation and pinpointing reliable noise for subsequent stages. Accordingly, we adopt different lightweight tuning approaches with effectiveness and efficiency in the following stages to further attenuate the influence of noise and alleviate the semantic gap among multi-typed behaviors. Extensive experiments on two real-world datasets demonstrate the superiority of DPT over a wide range of state-of-the-art methods. The implementation code is available online at https://github.com/zc-97/DPT.

The Industrial Internet of Things (IIoT) plays an important role in the digital management process of industrial enterprises. It improves the productivity of enterprises and reduces the management cost of enterprises. Anomaly detection has gradually become an important tool in the field of IIoT security. Due to the complex topology network and randomness of IIoT, the increase in data dimension and information distortion makes it difficult to extract data neighborhood information effectively. This is a challenge for current detection methods. In this article, a new multivariate time-series (MTSs) anomaly detection framework is proposed. The framework involves an intuition-based neutrosophic representation method for MTS and an automatic learning graph structure. First, the data preprocessing and the data representation are performed. Then, the fuzzy features of the MTS data after dimensionality reduction and co-frequency processing are captured by using the intuition-based neutrosophic model. Finally, the graph model is used to identify rare events in the data and interpret the sensor-level anomalies. Experiments on two public and available anomaly detection data sets prove that our method has better robustness and interpretability in the anomaly detection tasks of IIoT.

Trajectory data collection is a common task with many applications in our daily lives. Analyzing trajectory data enables service providers to enhance their services, which ultimately benefits users. However, directly collecting trajectory data may give rise to privacy-related issues that cannot be ignored. Local differential privacy (LDP), as the de facto privacy protection standard in a decentralized setting, enables users to perturb their trajectories locally and provides a provable privacy guarantee. Existing approaches to private trajectory data collection in a local setting typically use relaxed versions of LDP, which cannot provide a strict privacy guarantee, or require some external knowledge that is impractical to obtain and update in a timely manner. To tackle these problems, we propose a novel trajectory perturbation mechanism that relies solely on an underlying location set and satisfies pure ε-LDP to provide a stringent privacy guarantee. In the proposed mechanism, each point's adjacent direction information in the trajectory is used in its perturbation process. Such information serves as an effective clue to connect neighboring points and can be used to restrict the possible region of a perturbed point in order to enhance utility. To the best of our knowledge, our study is the first to use direction information for trajectory perturbation under LDP. Furthermore, based on this mechanism, we present an anchor-based method that adaptively restricts the region of each perturbed trajectory, thereby significantly boosting performance without violating the privacy constraint. Extensive experiments on both real-world and synthetic datasets demonstrate the effectiveness of the proposed mechanisms.

Sequential recommender systems aim to recommend the next items in which target users are most interested based on their historical interaction sequences. In practice, historical sequences typically contain some inherent noise (e.g., accidental interactions), which is harmful to learn accurate sequence representations and thus misleads the next-item recommendation. However, the absence of supervised signals (i.e., labels indicating noisy items) makes the problem of sequence denoising rather challenging. To this end, we propose a novel sequence denoising paradigm for sequential recommendation by learning hierarchical item inconsistency signals. More specifically, we design a hierarchical sequence denoising (HSD) model, which first learns two levels of inconsistency signals in input sequences, and then generates noiseless subsequences (i.e., dropping inherent noisy items) for subsequent sequential recommenders. It is noteworthy that HSD is flexible to accommodate supervised item signals, if any, and can be seamlessly integrated with most existing sequential recommendation models to boost their performance. Extensive experiments on five public benchmark datasets demonstrate the superiority of HSD over state-of-the-art denoising methods and its applicability over a wide variety of mainstream sequential recommendation models. The implementation code is available at https://github.com/zc-97/HSD

Accurate air quality forecasting is essential in managing outdoor activity risk and responding to pollution emergencies. However, effectively modeling complex underlying spatiotemporal dependencies among monitoring stations remains a challenging task. Most existing methods deeply rely on local features to model dynamic spatial correlations and on RNNs to model temporal evolution. In this paper, we propose a novel multi-task deep spatiotemporal graph neural network, named MTGnet, for air quality prediction. MTGnet's main advantage is its ability to adaptively capture complex correlations among different stations, represented as an adjacency matrix, using both local features and global patterns. MTGnet consists of multiple convolutional layers for aggregating information about nearby stations and extracting essential spatial and temporal features for future air quality prediction. Using this architecture, we implement a multi-task learning scheme that trains the model to predict air quality both finely, at the station level, and coarsely, at the city level. Experiments on multiple real datasets demonstrate that MTGnet outperforms state-of-the-art methods.

As an emerging paradigm, session-based recommendation is aimed at recommending the next item based on a set of anonymous sessions. Effectively representing a session that is normally a short interaction sequence renders a major technical challenge. In view of the limitations of pioneering studies that explore collaborative information from other sessions, in this paper we propose a new direction to enhance session representations by learning multi-faceted session-independent global item relations. In particular, we identify three types of advantageous global item relations, including negative relations that have not been studied before, and propose different graph construction methods to capture such relations. We then devise a novel multi-faceted global item relation (MGIR) model to encode different relations using different aggregation layers and generate enhanced session representations by fusing positive and negative relations. Our solution is flexible to accommodate new item relations and can easily integrate existing session representation learning methods to generate better representations from global relation enhanced session information. Extensive experiments on three benchmark datasets demonstrate the superiority of our model over a large number of state-of-the-art methods. Specifically, we show that learning negative relations is critical for session-based recommendation.

Social Internet of Things (SIoT) incorporates social relationship into the Internet of Things (IoT), and a compositive relationship between persons, devices, and persons to devices is utilized for providing better services. This article proposes a novel type of search, namely, top-$k$social spatial keyword search (SSKS) in SIoT networks to discover relevant users or data objects according to social, spatial, and textual preferences. Existing works mainly focus on two of these preferences at the same time, and efficiently processing top-$k$SSKS remains challenging. To this end, we propose two algorithms to evaluate top-$k$SSKS in SIoT networks. The first algorithm is a forward search-based algorithm, which spreads the search from the vertex of the querying user. An effective pruning strategy is established by recognizing an early termination condition according to the threefold preference. The forward search-based algorithm is efficient when textual objects are dense. The second algorithm is based on index searching. We present an index namely 2HL-GIL to support spatial and textual pruning while providing fast computation of social distances in the SIoT. Then an index-based search algorithm is proposed for top-$k$SSKS, and it is efficient especially when textual objects are sparse. Our proposed algorithms are evaluated over two real-life social networks attached with synthetic locations and textual data. Evaluation results illustrate the effectiveness and efficiency of our proposed forward search-based algorithm and index-based search algorithm.

In this paper, we study the problem of fine-grained air quality inference that predicts the air quality level of any location from air quality readings of nearby monitoring stations. We point out the importance of explicitly modeling both static and dynamic spatial correlations, and consequently propose a novel multi-channel attention model (MCAM) that models static and dynamic spatial correlations as separate channels. The static channel combines the beauty of attention mechanisms and graph-based spatial modeling via an adapted bilateral filtering technique, which considers not only locations' Euclidean distances but also their similarity of geo-context features. The dynamic channel learns stations' time-dependent spatial influence on a target location at each time step via long short-term memory (LSTM) networks and attention mechanisms. In addition, we introduce two novel ideas, atmospheric dispersion theories and the hysteretic nature of air pollutant dispersion, to better model the dynamic spatial correlation. We also devise a multi-channel graph convolutional fusion network to effectively fuse the graph outputs, along with other features, from both channels. Our extensive experiments on real-world benchmark datasets demonstrate that MCAM significantly outperforms the state-of-the-art solutions.

The ubiquity of devices in Internet of Things (IoT) has opened up a large source for IoT data. Machine learning (ML) models with big IoT data is beneficial to our daily life in monitoring air condition, pollution, climate change, etc. However, centralized conventional ML models rely on all clients' data at a central server, which seriously threatens user privacy. Federated learning (FL) emerges as a promising solution aiming to protect user privacy by enabling model training on a large corpus of decentralized data. The recent studies indicate FL suffers from the heterogeneity issue as it treats all clients' data equally, that is, FL might sacrifice the performance of the majority of clients to accommodate the performance of the minority of clients with low usability data. In order to overcome this issue, a reinforcement learning (RL)-based intelligent central server with the capability of recognizing heterogeneity is implemented, which can help lead the trend toward better performance for majority of clients. To be specific, an FL central server analyses the benefits of different collaboration by capturing the intricate patterns in heterogeneous clients based on rating feedback and then updates clients' weights iteratively, until it establishes a coalition of clients with quasioptimal performance. The experimental results on three real data sets under various heterogeneity levels demonstrate the superior performance of the proposed solution.

Computer vision and deep neural networks have been significantly promoting the development of visual perception in these years. Particularly, for autonomous vehicles, real-time image/video data is captured by onboard cameras and analyzed by computer vision techniques in many real applications. In the captured camera data, some contents can be used as auxiliary information to infer individuals' locations and trajectories, which leads to severe privacy leakage but has been rarely studied. Thus, the goal of this article is to protect individuals' location privacy by hiding side-channel information in the captured data while preserving the data utility for downstream applications. To this end, the technology of generative adversarial networks (GAN) is utilized to design two novel models, named ADGAN-I and ADGAN-II, both of which can take the original camera data as inputs and generate privacy-preserving outputs according to predefined sensitive object class. Thus, the processed camera data can defend location inference attack from adversaries in offline applications. Moreover, in ADGAN-I and ADGAN-II, the tradeoff between location privacy and data utility can be effectively balanced. Finally, the results of extensive real-data experiments validate the superiority of our proposed models over the state of the arts in utility preservation and privacy protection for autonomous vehicles' images and videos.

The rapid growth of DNA-sequencing technologies motivates more personalized and predictive genetic-oriented services, which further attract individuals to increasingly release their genome information to learn about personalized medicines, disease predispositions, genetic compatibilities, etc. Individual genome information is notoriously privacy-sensitive and highly associated with relatives. In this paper, we present an inference attack algorithm to predict target genotypes and phenotypes based on belief propagation in factor graphs. With this algorithm, an attacker can effectively predict the target genotypes and phenotypes of target individuals based on genome information shared by individuals or their relatives, and genotype and phenotype association from genome-wide association study (GWAS). To address the privacy threats resulted from such inference attacks, we elaborate the metrics to evaluate data utility and privacy and then present a data sanitization method. We evaluate our inference attack algorithm and data sanitization method on real GWAS dataset: Age-related macular degeneration (AMD) case/control dataset. The evaluation results show that our work can effectively defense against genome threats while guaranteeing data utility.

The autonomous driving (auto-driving) technology has been promoted significantly by the rapid advances in computer vision and deep neural networks. Auto-driving vehicles, nowadays, are fully equipped with numerous sensors such as cameras, geo-sensors, and radar sensors, to capture real-time data inside the vehicles and outside surroundings. Meanwhile, the captured data contains lots of private information about vehicles, drivers and passengers and thus faces a high risk of privacy breaches. Especially, side-channel information can be mined from camera data to identify vehicles' locations and even trajectories, raising serious privacy issues. Unfortunately, the issue, how to resist location-inference attack for camera data in auto-driving, has never been addressed in literature. In this paper, we intend to fill this blank by developing a GAN-based image-toimage translation method named Auto-Driving GAN (ADGAN). Through performance comparisons between ADGAN and the state-of-the-art, the superiority of ADGAN can be validated - offering an effective tradeoff between recognition utility and privacy protection for camera data.

Location-based services have become a mainstream in people’s daily lives due to continuous innovations in the field of mobile networking and GPS technologies. Recently they have advanced into a hot topic to which the majority of researchers pay close attention about how to enjoy them while safeguarding the location privacy of mobile users. Existing works involve the injection of random noise that cannot pledge the quality of service. Herein this manuscript, we propose a novel location privacy protection model based on the loss of service quality. This model allows the user to express his/her requirement of service quality by specifying the maximum service quality loss Lmax , which is the user’s tolerance. Lmax can be set to 0. Our comprehensive experimental evaluation using a real-world dataset demonstrates that our modus outdoes other state-of-the-art approaches.

A number of security and privacy challenges of cyber system are arising due to the rapidly evolving scale and complexity of modern system and networks. The cyber system is a fundamental ingredient for Internet of Things (IoT) and smart city which are driven by huge amount of data. These data carry a lot of information for mining and analysis, especially trajectory data. If unprotected trajectory data is released, it may disclose user’s personal privacy, such as home, religion, and behavior mode, which will endanger their personal security. Until now, many methods for protecting trajectory information have been proposed. However, these methods have the following deficiencies: (i) they cannot defend against speculative attacks if the attacker’s background knowledge is maximized; (ii) when studying the problem, they made some strong assumptions that did not match the reality; (iii) the implementation algorithm is complicated and the time complexity is high, which means that data cannot be executed quickly when the amount is large. So, in this paper, we propose a spatial partition based method to publish trajectory data via differential privacy. First, by exponential mechanism, we divide location set at the same time into different partitions fast and accurately. Then we propose another effective method to release trajectory in a differential private manner. We design experiment based on the real-life dataset and compare it with existing method. The results show that the trajectory dataset released by our algorithm has better usability while ensuring privacy.

Mobile devices bring benefits as well as the risk of exposing users’ location information, as some embedded sensors can be accessed without users’ permission and awareness. In this paper, we show that, only by using the data collected from the embedded sensors in mobile devices instead of GPS data, we can infer a user’s location information with high accuracy. Three issues are addressed which are route identification, user localization in a specific route, and user localization in a bounded area. The Dynamic Time Warping based technique is designed and we develop a Hidden Markov Model to solve the localization problem. Real experiments are performed to evaluate our proposed methods.

Identification of ideal mid-sagittal line (iML) is important for image registration, brain segmentation, pathology detection and particularly for medical image classification. In this paper, iML extraction method based on scale invariant feature transform (SIFT) features is proposed for brain CT images. The method consists of an offline part and an online part. In the offline part, the iML feature points of training set is extracted by an auxiliary tool and an optimized matching template set is obtained by our feature fusion and filtering algorithms. In the online part, a matching point set is generated by matching SIFT features of test images to the offline template. Then the point set is refined by our pruning algorithm and iMLs of test images are fitted by the refined point set. Both real and simulated image data sets are used to verify the accuracy, robustness and execution efficiency of the algorithm. Experimental results show that, our method achieves good accuracy and efficiency in both real and simulation image sets, and performs better tolerance to rotation, noise, fuzzy and asymmetry in comparison with other existing algorithms.

In this work, an interactive visual system MICS is presented for large-scale brain CT image classification. Automatic feature extraction algorithms are added in MICS to improve system efficiency and classification accuracy. In visualization part, we designed an interactive feature extraction interface, enable users to extract and fine-tune image features according to specific requirements. In addition, all image features in database are visualized as dynamic charts in every phase of classification. These allow users to compare the current image with others in some specific feature and re-mark the possible misclassification. Finally, by series experiments and case studies, we verify the performance of the classification algorithm as well as the effectiveness and applicability of the visual design in MICS.

Abstract Histograms are important tools for data mining and analysis. Several differentially private publishing schemes for histograms have been proposed recently. Existing differentially private histogram publication schemes have shown that histogram reconstruction is a promising idea for the improvement of publication histograms' accuracy. However, none of these have properly considered the problem outliers in the original histogram, which can cause significant reconstruction errors. Based on the problem, the publication of histogram outliers under differential privacy, this paper puts forward a publication method for histograms with outliers under differential privacy: Outlier‐HistoPub. Our method deals with the count sequence of the original histogram first, using a “global sort” to reduce the degree of alternative distribution (a concept proposed in this paper), which may eliminate the influence of outliers during reconstruction. To avoid individual privacy leakage in the reconstruction process, an exponential mechanism is used to select the most similar adjacent bins of the uniformity distribution histogram to merge each time, and the Laplace mechanism is utilized to generate noisy data to perturb the count sequence of the reconstruction histogram. Experiments prove that the method proposed in this paper can improve the efficiency and accuracy of histogram publication. Copyright © 2016 John Wiley & Sons, Ltd.

Medical images are one of the most important tools in doctors' diagnostic decision-making. It has been a research hotspot in medical big data that how to effectively represent medical images and find essential patterns hidden in them to assist doctors to achieve a better diagnosis. Several graph models have been developed to represent medical images. However, the unique structures of domain-specific images are not considered well to lose some essential information. Thus, aiming at brain CT images, we first construct a graph about the Topological Relations between Ventricles and Lesions (TRVL) and present the graph modeling process. Then we propose a method named Frequent Approximate Subgraph Mining based on Graph Edit Distance (FASMGED). This method uses an error-tolerant graph matching strategy that is accordant with ubiquitous noise in practice. Experimental results show that the graph modeling process is computationally scalable and FASMGED can find more significant patterns than current algorithms.

A number of brain computed tomography (CT) images stored in hospitals that contain valuable information should be shared to support computer-aided diagnosis systems. Finding the similar brain CT images from the brain CT image database can effectively help doctors diagnose based on the earlier cases. However, the similarity retrieval for brain CT images requires much higher accuracy than the general images. In this paper, a new model of uncertain location graph (ULG) is presented for brain CT image modeling and similarity retrieval. According to the characteristics of brain CT image, we propose a novel method to model brain CT image to ULG based on brain CT image texture. Then, a scheme for ULG similarity retrieval is introduced. Furthermore, an effective index structure is applied to reduce the searching time. Experimental results reveal that our method functions well on brain CT images similarity retrieval with higher accuracy and efficiency.