Pretrained vision-language models exhibit strong zero-shot classification capabilities, but their predictions degrade significantly under common image corruptions. To improve robustness, many test-time adaptation (TTA) methods adopt positive data augmentation (PDA), which generates multiple views of each test sample to reduce prediction variance. However, these methods suffer from two key limitations. First, it introduces considerable computational overhead due to the large number of augmentations required per image. Second, it fails to mitigate prediction bias, where the model tends to predict certain classes disproportionately under corruption, as PDA operates on corrupted inputs and typically does not remove the corruption itself. To address these challenges, we propose Panda, a novel TTA method based on negative data augmentation (NDA). Unlike positive augmentations that preserve object semantics, Panda generates negative augmentations by disrupting semantic content. It divides images into patches and randomly assembles them from a shared patch pool. These negatively augmented images retain corruption-specific features while discarding object-relevant signals. We then subtract the mean feature of these negative samples from the original image feature, effectively suppressing corruption-related components while preserving class-relevant information. This mitigates prediction bias under distribution shifts. Importantly, Panda allows augmentation to be shared across samples within a batch, resulting in minimal computational overhead. Panda can be seamlessly integrated into existing test-time adaptation frameworks and substantially improve their robustness. Our experiments indicate that Panda delivers superior performance compared to PDA methods, and a wide range of TTA methods exhibit significantly enhanced performance when integrated with Panda.

Yuanchen Bei, Tianxin Wei, Xuying Ning, Yanjun Zhao, Zhining Liu, Xiao Lin, Yada Zhu, Hendrik Hamann, Jingrui He, Hanghang Tong. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Chengming Cui, Tianxin Wei, Ziyi Chen, Ruizhong Qiu, Zhichen Zeng, Zhining Liu, Xuying Ning, Duo Zhou, Jingrui He. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Haobo Xu, Sirui Chen, Ruizhong Qiu, Yuchen Yan, Chen Luo, Monica Xiao Cheng, Jingrui He, Hanghang Tong. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Modern power systems are becoming increasingly dynamic, with changing topologies and time-varying loads driven by renewable energy variability, electric vehicle adoption, and active grid reconfiguration. Despite these changes, publicly available test cases remain scarce, due to security concerns and the significant effort required to anonymize real systems. Such limitations call for generative tools that can jointly synthesize grid structure and nodal dynamics. However, modeling the joint distribution of network topology, branch attributes, bus properties, and dynamic load profiles remains a major challenge, while preserving physical feasibility and avoiding prohibitive computational costs. We present PowerGrow, a co-generative framework that significantly reduces computational overhead while maintaining operational validity. The core idea is dependence decomposition: the complex joint distribution is factorized into a chain of conditional distributions over feasible grid topologies, time-series bus loads, and other system attributes, leveraging their mutual dependencies. By constraining the generation process at each stage, we implement a hierarchical graph beta-diffusion process for structural synthesis, paired with a temporal autoencoder that embeds time-series data into a compact latent space, improving both training stability and sample fidelity. Experiments across benchmark settings show that PowerGrow not only outperforms prior diffusion models in fidelity and diversity but also achieves a 98.9% power flow convergence rate and improved N-1 contingency resilience. This demonstrates its ability to generate operationally valid and realistic power grid scenarios.

Federated cross-domain recommendation (Federated CDR) aims to collaboratively learn personalized recommendation models across heterogeneous domains while preserving data privacy. Recently, large language model (LLM)-based recommendation models have demonstrated impressive performance by leveraging LLMs' strong reasoning capabilities and broad knowledge. However, adopting LLM-based recommendation models in Federated CDR scenarios introduces new challenges. First, there exists a risk of overfitting with domain-specific local adapters. The magnitudes of locally optimized parameter updates often vary across domains, causing biased aggregation and overfitting toward domain-specific distributions. Second, unlike traditional recommendation models (e.g., collaborative filtering, bipartite graph-based methods) that learn explicit and comparable user/item representations, LLMs encode knowledge implicitly through autoregressive text generation training. This poses additional challenges for effectively measuring the cross-domain similarities under heterogeneity. To address these challenges, we propose an LLM-based framework for federated cross-domain recommendation, FeDecider. Specifically, FeDecider tackles the challenge of scale-specific noise by disentangling each client's low-rank updates and sharing only their directional components. To handle the need for flexible and effective integration, each client further learns personalized weights that achieve the data-aware integration of updates from other domains. Extensive experiments across diverse datasets validate the effectiveness of our proposed FeDecider.

Zihao Li, Lecheng Zheng, Bowen Jin, Dongqi Fu, Baoyu Jing, Yikun Ban, Jingrui He, Jiawei Han. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

In the era of foundation models, Out-of-Distribution (OOD) problems, i.e., the data discrepancy between the training environments and testing environments, hinder AI generalization. Further, relational data like graphs disobeying the Independent and Identically Distributed (IID) condition makes the problem more challenging, especially much harder when it is associated with time. Motivated by this, to realize the robust invariant learning over temporal graphs, we want to investigate what components in temporal graphs are most invariant and representative with respect to labels. With the Information Bottleneck (IB) method, we propose an error-bounded Invariant Link Selector that can distinguish invariant components and variant components during the training process to make the deep learning model generalizable for different testing scenarios. Besides deriving a series of rigorous generalizable optimization functions, we also equip the training with task-specific loss functions, e.g., temporal link prediction, to make pre-trained models solve real-world application tasks like citation recommendation and merchandise recommendation, as demonstrated in our experiments with state-of-the-art (SOTA) methods. Our code is available at \url{https://github.com/kthrn22/OOD-Linker}

Climate science studies the structure and dynamics of Earth's climate system and seeks to understand how climate changes over time, where the data is usually stored in the format of time series, recording the climate features, geolocation, time attributes, etc. Recently, much research attention has been paid to the climate benchmarks. In addition to the most common task of weather forecasting, several pioneering benchmark works are proposed for extending the modality, such as domain-specific applications like tropical cyclone intensity prediction and flash flood damage estimation, or climate statement and confidence level in the format of natural language. To further motivate the artificial intelligence development for climate science, in this paper, we first contribute a multi-modal climate benchmark, i.e., ClimateBench-M, which aligns (1) the time series climate data from ERA5, (2) extreme weather events data from NOAA, and (3) satellite image data from NASA HLS based on a unified spatial-temporal granularity. Second, under each data modality, we also propose a simple but strong generative method that could produce competitive performance in weather forecasting, thunderstorm alerts, and crop segmentation tasks in the proposed ClimateBench-M. The data and code of ClimateBench-M are publicly available at https://github.com/iDEA-iSAIL-Lab-UIUC/ClimateBench-M.

Graph Self-Supervised Learning (SSL) has emerged as a pivotal area of research in recent years. By engaging in pretext tasks to learn the intricate topological structures and properties of graphs using unlabeled data, these graph SSL models achieve enhanced performance, improved generalization, and heightened robustness. Despite the remarkable achievements of these graph SSL methods, their current implementation poses significant challenges for beginners and practitioners due to the complex nature of graph structures, inconsistent evaluation metrics, and concerns regarding reproducibility hinder further progress in this field. Recognizing the growing interest within the research community, there is an urgent need for a comprehensive, beginner-friendly, and accessible toolkit consisting of the most representative graph SSL algorithms. To address these challenges, we present a Graph SSL toolkit named PyG-SSL, which is built upon PyTorch and is compatible with various deep learning and scientific computing backends. Within the toolkit, we offer a unified framework encompassing dataset loading, hyper-parameter configuration, model training, and comprehensive performance evaluation for diverse downstream tasks. Moreover, we provide beginner-friendly tutorials and the best hyper-parameters of each graph SSL algorithm on different graph datasets, facilitating the reproduction of results. The GitHub repository of the library is https://github.com/iDEA-iSAIL-Lab-UIUC/pyg-ssl.

Generative Adversarial Network approaches such as StyleGAN and StyleGAN2 provide two key benefits: the ability to generate photo-realistic face images and possessing a semantically structured latent space from which these images are created. Many approaches have emerged for editing images derived from vectors in the latent space of a pre-trained StyleGAN and StyleGAN2 models by identifying semantically meaningful directions (e.g., gender, age or hair color) in the latent space. By moving the vector in a specific direction, the ideal result would only change the target feature while preserving all the other features. Providing an ideal data augmentation approach for gesture research as it could be used to generate numerous image variations whilst keeping the facial gestures and expressions intact. However, the entanglement issue, where changing one feature inevitably affects other features, impacts the ability to preserve facial gestures and expressions. To address the entanglement issue, we propose the use of an addition to the loss function of a Facial Keypoint Detection model to restrict changes to the facial gesture. Our approach builds on top of an existing model and simply adds the proposed Human Face Landmark Detection (HFLD) loss, provided by a pre-trained Facial Keypoint Detection model, to the original loss function. We quantitatively and qualitatively evaluate the existing and our extended model, showing the effectiveness of our proposed loss term in addressing the entanglement issue and maintaining the facial gesture and expression. Our approach achieved up to 49% reduction in the change of emotion in our experiments. Moreover, we show the benefit of our approach by comparing with state-of-the-art models. By increasing the ability to preserve the facial gesture and expression during facial transformation, we present a way to create human face images with fixed expression but different appearances, making it a reliable data augmentation approach for Facial Gesture and Expression research.

Dataset Distillation (DD) aims to generate a compact synthetic dataset that enables models to achieve performance comparable to training on the full large dataset, significantly reducing computational costs. Drawing from optimal transport theory, we introduce WMDD (Wasserstein Metric-based Dataset Distillation), a straightforward yet powerful method that employs the Wasserstein metric to enhance distribution matching. We compute the Wasserstein barycenter of features from a pretrained classifier to capture essential characteristics of the original data distribution. By optimizing synthetic data to align with this barycenter in feature space and leveraging per-class BatchNorm statistics to preserve intra-class variations, WMDD maintains the efficiency of distribution matching approaches while achieving state-of-the-art results across various high-resolution datasets. Our extensive experiments demonstrate WMDD's effectiveness and adaptability, highlighting its potential for advancing machine learning applications at scale.

Nowadays, temporal heterogeneous graphs attract much research and industrial attention for building the next-generation Relational Deep Learning models and applications, due to their informative structures and features. While providing timely and precise services like personalized recommendations and question answering, this rich information also introduces extra exposure risk for each node in the graph. The distinctive local topology, the abundant heterogeneous features, and the time dimension of the graph data are more prone to expose sensitive information and narrow down the scope of victim candidates, which calls for well-defined protection techniques on graphs. To this end, we propose a Temporal Heterogeneous Graph Generator balancing Privacy, Utility, and Efficiency, named THePUff. More specifically, we first propose a differential privacy algorithm to perturb the input temporal heterogeneous graph for protecting privacy, and then utilize both the perturbed graph and the original one in a generative adversarial setting for THePUff to learn and generate privacy-guaranteed and utility-preserved graph data in an efficient manner. We further propose 6 new metrics in the temporal setting to measure heterogeneous graph utility and privacy. Finally, based on temporal heterogeneous graph datasets with up to 1 million nodes and 20 million edges, the experiments show that THePUff generates utilizable temporal heterogeneous graphs with privacy protected, compared with state-of-the-art baselines.

Powerful as they are, graph neural networks (GNNs) are known to be vulnerable to distribution shifts. Recently, test-time adaptation (TTA) has attracted attention due to its ability to adapt a pre-trained model to a target domain, without re-accessing the source domain. However, existing TTA algorithms are primarily designed for attribute shifts in vision tasks, where samples are independent. These methods perform poorly on graph data that experience structure shifts, where node connectivity differs between source and target graphs. We attribute this performance gap to the distinct impact of node attribute shifts versus graph structure shifts: the latter significantly degrades the quality of node representations and blurs the boundaries between different node categories. To address structure shifts in graphs, we propose Matcha, an innovative framework designed for effective and efficient adaptation to structure shifts by adjusting the htop-aggregation parameters in GNNs. To enhance the representation quality, we design a prediction-informed clustering loss to encourage the formation of distinct clusters for different node categories. Additionally, Matcha seamlessly integrates with existing TTA algorithms, allowing it to handle attribute shifts effectively while improving overall performance under combined structure and attribute shifts. We validate the effectiveness of Matcha on both synthetic and real-world datasets, demonstrating its robustness across various combinations of structure and attribute shifts. Our code is available at https://github.com/baowenxuan/Matcha.

Time-series forecasting plays a critical role in many real-world applications. Although increasingly powerful models have been developed and achieved superior results on benchmark datasets, through a fine-grained sample-level inspection, we find that (i) no single model consistently outperforms others across different test samples, but instead (ii) each model excels in specific cases. These findings prompt us to explore how to adaptively leverage the distinct strengths of various forecasting models for different samples. We introduce TimeFuse, a framework for collective time-series forecasting with sample-level adaptive fusion of heterogeneous models. TimeFuse utilizes meta-features to characterize input time series and trains a learnable fusor to predict optimal model fusion weights for any given input. The fusor can leverage samples from diverse datasets for joint training, allowing it to adapt to a wide variety of temporal patterns and thus generalize to new inputs, even from unseen datasets. Extensive experiments demonstrate the effectiveness of TimeFuse in various long-/short-term forecasting tasks, achieving near-universal improvement over the state-of-the-art individual models. Code is available at https://github.com/ZhiningLiu1998/TimeFuse.

Real-world multimodal data usually exhibit complex structural relationships beyond traditional one-to-one mappings like image-caption pairs. Entities across modalities interact in intricate ways, with images and text forming diverse interconnections through contextual dependencies and co-references. Graphs provide powerful structural information for modeling intra-modal and inter-modal relationships. However, previous works fail to distinguish multi-hop neighbors and treat the graph as a standalone modality, which fragments the overall understanding. This limitation presents two key challenges in multimodal learning: (1) integrating structural information from multi-hop neighbors into foundational models, and (2) fusing modality-specific information in a principled manner. To address these challenges, we revisit the role of graphs in multimodal learning within the era of foundation models and propose Graph4MM, a graph-based multimodal learning framework. To be specific, we introduce Hop-Diffused Attention, which integrates multi-hop structural information into self-attention through causal masking and hop diffusion. Furthermore, we design MM-QFormer, a multi-mapping querying transformer for cross-modal fusion. Through theoretical and empirical analysis, we show that leveraging structures to integrate both intra- and inter-modal interactions improves multimodal understanding beyond treating them as a standalone modality. Experiments on both generative and discriminative tasks show that Graph4MM outperforms larger VLMs, LLMs, and multimodal graph baselines, achieving a 6.93% average improvement.

Accurate predictions rely on the expressiveness power of graph deep learning frameworks like graph neural networks and graph transformers, where a positional encoding mechanism has become much more indispensable in recent state-of-the-art (SOTA) works to record the canonical position information. However, the current positional encoding limits in three aspects, at least: (1) most positional encodings are pre-defined, and fixed functions, which are inadequate to adapt to the complex attributed graphs; (2) a few pioneering works propose the learnable positional encoding but still limited to the structural information, leaving the real-world time-evolving topological and feature information untouched; (3) most positional encodings should be equipped with transformer's attention mechanism to fully release the power, where the dense or relational attention is often unaffordable on large-scale structured data. Hence, we study the possibility of Learnable Spatial-Temporal Positional Encoding in an effective and efficient manner and then propose a simple temporal link prediction model named L-STEP. Briefly, for L-STEP, we (1) prove the proposed positional learning scheme can preserve the graph property from the spatial-temporal spectral viewpoint, (2) verify that MLPs can fully exploit the expressiveness and reach Transformers' performance on that encoding, (3) change different initial positional encoding inputs to show robustness, (4) analyze the theoretical complexity and obtain less empirical running time than SOTA, and (5) demonstrate its temporal link prediction out-performance on 13 classic datasets and with 10 algorithms in both transductive and inductive settings using 3 different sampling strategies. Also, L-STEP obtains the leading performance in the newest large-scale TGB benchmark.

Mixture-of-Experts (MoE) Transformer, the backbone architecture of multiple phenomenal language models, leverages sparsity by activating only a fraction of model parameters for each input token. The sparse structure, while allowing constant time costs, results in space inefficiency: we still need to load all the model parameters during inference. We introduce ResMoE, an innovative MoE approximation framework that utilizes Wasserstein barycenter to extract a common expert (barycenter expert) and approximate the residuals between this barycenter expert and the original ones. ResMoE enhances the space efficiency for inference of large-scale MoE Transformers in a one-shot and data-agnostic manner without retraining while maintaining minimal accuracy loss, thereby paving the way for broader accessibility to large language models. We demonstrate the effectiveness of ResMoE through extensive experiments on Switch Transformer, Mixtral, and DeepSeekMoE models. The results show that ResMoE can reduce the number of parameters in an expert by up to 75% while maintaining comparable performance. The code is available at https://github.com/iDEA-iSAIL-Lab-UIUC/ResMoE, and the supplementary appendix is available at https://famous-blue-raincoat.github.io/mengtingai/files/ResMoE_Appendix.pdf.

Knowledge graphs (KGs), which store an extensive number of relational facts, serve various applications. Recently, personalized knowledge graphs (PKGs) have emerged as a solution to optimize storage costs by customizing their content to align with users' specific interests within particular domains. In the real world, on the one hand, user queries and their underlying interests are inherently evolving, requiring PKGs to adapt continuously; on the other hand, the summarization is constantly expected to be as small as possible in terms of storage cost. However, the existing PKG summarization methods implicitly assume that the user's interests are constant and do not shift. Furthermore, when the size constraint of PKG is extremely small, the existing methods cannot distinguish which facts are more of immediate interest and guarantee the utility of the summarized PKG. To address these limitations, we propose APEX2, a highly scalable PKG summarization framework designed with robust theoretical guarantees to excel in adaptive summarization tasks with extremely small size constraints. To be specific, after constructing an initial PKG, APEX2 continuously tracks the interest shift and adjusts the previous summary. The experiments show that APEX outperforms state-of-the-art baselines in terms of both query-answering accuracy and efficiency.

Distribution shifts between training and testing samples frequently occur in practice and impede model generalization performance. This crucial challenge thereby motivates studies on domain generalization (DG), which aim to predict the label on unseen target domain data by solely using data from source domains. It is intuitive to conceive the class-separated representations learned in contrastive learning (CL) are able to improve DG, while the reality is quite the opposite: users observe directly applying CL deteriorates the performance. We analyze the phenomenon with the insights from CL theory and discover lack of intra-class connectivity in the DG setting causes the deficiency. We thus propose a new paradigm, domain-connecting contrastive learning (DCCL), to enhance the conceptual connectivity across domains and obtain generalizable representations for DG. On the data side, more aggressive data augmentation and cross-domain positive samples are introduced to improve intra-class connectivity. On the model side, to better embed the unseen test domains, we propose model anchoring to exploit the intra-class connectivity in pre-trained representations and complement the anchoring with generative transformation loss. Extensive experiments on five standard DG benchmarks are performed. The results verify that DCCL outperforms state-of-the-art baselines even without domain supervision. The detailed model implementation and the code are provided through https://github.com/weitianxin/DCCL.

Pretrained vision-language models such as CLIP achieve strong zero-shot generalization but remain vulnerable to distribution shifts caused by input corruptions. In this work, we investigate how corruptions affect CLIP’s image embeddings and uncover a consistent phenomenon we term as embedding variance collapse, where both intra-class and inter-class variances shrink as corruption severity increases. We find that this collapse is closely tied to performance degradation, with inter-class variance strongly correlated with classification accuracy. To explain this phenomenon, we analyze how corruptions alter the structure of the embedding space. Our theoretical results suggest that the visual encoder tends to encode corruption-related signals, which dilute class-discriminative features and compress the representation geometry. We further show that maximizing inter-class variance, even when estimated from pseudo-labels, can provably enhance embedding quality. Based on this insight, we propose Mint, a simple test-time adaptation method that maximizes pseudo-label-based inter-class variance on the fly using a mean accumulator and a gradient accumulator. Mint operates effectively with small batch sizes and consistently improves performance across multiple corruption benchmarks and CLIP architectures. Our code is available at https://github.com/baowenxuan/Mint.

Class-imbalanced learning (CIL) on tabular data is important in many real-world applications where the minority class holds the critical but rare outcomes. In this paper, we present CLIMB, a comprehensive benchmark for class-imbalanced learning on tabular data. CLIMB includes 73 real-world datasets across diverse domains and imbalance levels, along with unified implementations of 29 representative CIL algorithms. Built on a high-quality open-source Python package with unified API designs, detailed documentation, and rigorous code quality controls, CLIMB supports easy implementation and comparison between different CIL algorithms. Through extensive experiments, we provide practical insights on method accuracy and efficiency, highlighting the limitations of naive rebalancing, the effectiveness of ensembles, and the importance of data quality. Our code, documentation, and examples are available at https://github.com/ZhiningLiu1998/imbalanced-ensemble.

Large language models are typically adapted to downstream tasks through supervised fine-tuning on domain-specific data. While standard fine-tuning focuses on minimizing generation loss to optimize model parameters, we take a deeper step by retaining and leveraging the model’s own learning signals, analogous to how human learners reflect on past mistakes to improve future performance. We first introduce the concept of Mistake Log to systematically track the model’s learning behavior and recurring errors throughout fine-tuning. Treating the original transformer-based model as the Pilot, we correspondingly design a Copilot model to refine the Pilot’s inference performance via logits rectification. We name the overall Pilot-Copilot framework the Transformer Copilot, which introduces (i) a novel Copilot model design, (ii) a joint training paradigm where the Copilot continuously learns from the evolving Mistake Log alongside the Pilot, and (iii) a fused inference paradigm where the Copilot rectifies the Pilot’s logits for enhanced generation. We provide both theoretical and empirical analyses on our new learning framework. Experiments on 12 benchmarks spanning commonsense, arithmetic, and recommendation tasks demonstrate that Transformer Copilot consistently improves performance by up to 34.5%, while introducing marginal computational overhead to Pilot models and exhibiting strong scalability and transferability. Our code is released at https://github.com/jiaruzouu/TransformerCopilot.

Process Reward Models (PRMs) have recently emerged as a powerful framework for supervising intermediate reasoning steps in large language models (LLMs). Previous PRMs are primarily trained on model final output responses and struggle to evaluate intermediate thinking trajectories robustly, especially in the emerging setting of trajectory–response outputs generated by frontier reasoning models like Deepseek-R1. In this work, we introduce ReasonFlux-PRM, a novel trajectory-aware PRM explicitly designed to evaluate the trajectory-response type of reasoning traces. ReasonFlux-PRM incorporates both step-level and trajectory-level supervision, enabling fine-grained reward assignment aligned with structured chain-of-thought data. We adapt ReasonFlux-PRM to support reward supervision under both offline and online settings, including (i) selecting high-quality model distillation data for downstream supervised fine-tuning of smaller models, (ii) providing dense process-level rewards for policy optimization during reinforcement learning, and (iii) enabling reward-guided Best-of-N test-time scaling. Empirical results on challenging downstream benchmarks such as AIME, MATH500, and GPQA-Diamond demonstrate that ReasonFlux-PRM-7B selects higher quality data than strong PRMs (e.g., Qwen2.5-Math-PRM-72B) and human-curated baselines. Furthermore, ReasonFlux-PRM-7B yields consistent performance improvements, achieving average gains of 12.1\% in supervised fine-tuning, 4.5\% in reinforcement learning, and 6.3\% in test-time scaling. We also release an efficient ReasonFlux-PRM-1.5B for resource-constrained applications and edge deployment. Our code and models are released at https://github.com/Gen-Verse/ReasonFlux.

Graph anomaly detection has gained significant attention across various domains, particularly in critical applications like fraud detection in e-commerce platforms and insider threat detection in cybersecurity.Usually, these data are composed of multiple types (e.g., user information and transaction records for financial data), thus exhibiting view heterogeneity.However, in the era of big data, the heterogeneity of views and the lack of label information pose substantial challenges to traditional approaches.Existing unsupervised graph anomaly detection methods often struggle with high-dimensionality issues, rely on strong assumptions about graph structures or fail to handle complex multi-view graphs.To address these challenges, we propose a cluster aware multi-view graph anomaly detection method, called CARE.Our approach captures both local and global node affinities by augmenting the graph's adjacency matrix with the pseudo-label (i.e., soft membership assignments) without any strong assumption about the graph.To mitigate potential biases from the pseudo-label, we introduce a similarity-guided loss.Theoretically, we show that the proposed similarity-guided loss is a variant of contrastive learning loss, and we present how this loss alleviates the bias introduced by pseudolabel with the connection to graph spectral clustering.Experimental results on several datasets demonstrate the effectiveness and efficiency of our proposed framework.Specifically, CARE outperforms the second-best competitors by more than 39% on the Amazon dataset with respect to AUPRC and 18.7% on the YelpChi dataset with respect to AUROC.The code of our method is available at the GitHub link: https://github.com/zhenglecheng/CARE-demo.

Transfer learning aims to transfer knowledge or information from a source domain to a relevant target domain. In this paper, we understand transfer learning from the perspectives of knowledge transferability and trustworthiness. This involves two research questions: How is knowledge transferability quantitatively measured and enhanced across domains? Can we trust the transferred knowledge in the transfer learning process? To answer these questions, this paper provides a comprehensive review of trustworthy transfer learning from various aspects, including problem definitions, theoretical analysis, empirical algorithms, and real-world applications. Specifically, we summarize recent theories and algorithms for understanding knowledge transferability under (within-domain) IID and non-IID assumptions. In addition to knowledge transferability, we review the impact of trustworthiness on transfer learning, e.g., whether the transferred knowledge is adversarially robust or algorithmically fair, how to transfer the knowledge under privacy-preserving constraints, etc. Beyond discussing the current advancements, we highlight the open questions and future directions for understanding transfer learning in a reliable and trustworthy manner.

With the prevalence of deep learning–based embedding approaches, recommender systems have become a proven and indispensable tool in various information filtering applications. However, it is difficult to determine which aspects of the deep models’ input drive the final ranking decision in many of these recommender systems; thus, they often cannot be understood by human stakeholders. In this article, we investigate the dilemma between recommendation and explainability, and show that by utilizing the contextual features (e.g., item reviews from users), we can design a series of explainable recommender systems without sacrificing their performance. In particular, we propose three types of explainable recommendation strategies with gradual changes in model transparency: whitebox, graybox, and blackbox. Each strategy explains its ranking decisions via different mechanisms: attention weights, adversarial perturbations, and counterfactual perturbations. We apply these explainable models on five real-world datasets under a contextualized setting in which the users provided explicit reviews of the items. The empirical results show that our models can achieve highly competitive ranking performance while generating accurate and effective explanations in terms of numerous quantitative metrics and qualitative visualizations.

In federated learning, most existing robust aggregation rules (AGRs) combat Byzantine attacks in the IID setting, where client data is assumed to be independent and identically distributed. In this paper, we address label skewness, a more realistic and challenging non-IID setting, where each client only has access to a few classes of data. In this setting, state-of-the-art AGRs suffer from selection bias, leading to significant performance drop for particular classes; they are also more vulnerable to Byzantine attacks due to the increased variation among gradients of honest clients. To address these limitations, we propose an efficient two-stage method named BOBA. Theoretically, we prove the convergence of BOBA with an error of the optimal order. Our empirical evaluations demonstrate BOBA’s superior unbiasedness and robustness across diverse models and datasets when compared to various baselines.

In recent years, Contrastive Learning (CL) has become a predominant representation learning paradigm for time series. Most existing methods manually build specific CL Strategies (CLS) by human heuristics for certain datasets and tasks. However, manually developing CLS usually requires excessive prior knowledge about the data, and massive experiments to determine the detailed CL configurations. In this paper, we present an Automated Machine Learning (AutoML) practice at Microsoft, which automatically learns CLS for time series datasets and tasks, namely Automated Contrastive Learning (AutoCL). We first construct a principled search space of size over 3 × 1012, covering data augmentation, embedding transformation, contrastive pair construction, and contrastive losses. Further, we introduce an efficient reinforcement learning algorithm, which optimizes CLS from the performance on the validation tasks, to obtain effective CLS within the space. Experimental results on various real-world datasets demonstrate that AutoCL could automatically find the suitable CLS for the given dataset and task. From the candidate CLS found by AutoCL on several public datasets/tasks, we compose a transferable Generally Good Strategy (GGS), which has a strong performance for other datasets. We also provide empirical analysis as a guide for the future design of CLS.

There have been tremendous efforts over the past decades dedicated to the generation of realistic graphs in a variety of domains, ranging from social networks to computer networks, from gene regulatory networks to online transaction networks. Despite the remarkable success, the vast majority of these works are unsupervised in nature and are typically trained to minimize the expected graph reconstruction loss, which would result in the representation disparity issue in the generated graphs, i.e., the protected groups (often minorities) contribute less to the objective and thus suffer from systematically higher errors. In this paper, we aim to tailor graph generation to downstream mining tasks by leveraging label information and user-preferred parity constraints. In particular, we start from the investigation of representation disparity in the context of graph generative models. To mitigate the disparity, we propose a fairness-aware graph generative model named Fairgen. Our model jointly trains a label-informed graph generation module and a fair representation learning module by progressively learning the behaviors of the protected and unprotected groups, from the ‘easy’ concepts to the ‘hard’ ones. In addition, we propose a generic context sampling strategy for graph generative models, which is proven to be capable of fairly capturing the contextual information of each group with a high probability. Experimental results on seven real-world data sets demonstrate that Fairgen (1) obtains performance on par with state-of-the-art graph generative models across nine network properties, (2) mitigates the representation disparity issues in the generated graphs, and (3) substantially boosts the model performance by up to 17% in downstream tasks via data augmentation.

We study both stream-based and pool-based active learning with neural network approximations. A recent line of works proposed bandit-based approaches that transformed active learning into a bandit problem, achieving both theoretical and empirical success. However, the performance and computational costs of these methods may be susceptible to the number of classes, denoted as $K$, due to this transformation. Therefore, this paper seeks to answer the question: "How can we mitigate the adverse impacts of $K$ while retaining the advantages of principled exploration and provable performance guarantees in active learning?" To tackle this challenge, we propose two algorithms based on the newly designed exploitation and exploration neural networks for stream-based and pool-based active learning. Subsequently, we provide theoretical performance guarantees for both algorithms in a non-parametric setting, demonstrating a slower error-growth rate concerning $K$ for the proposed approaches. We use extensive experiments to evaluate the proposed algorithms, which consistently outperform state-of-the-art baselines.

Recent works have shown a reduction from contextual bandits to online regression under a realizability assumption (Foster and Rakhlin, 2020; Foster and Krishnamurthy, 2021). In this work, we investigate the use of neural networks for such online regression and associated Neural Contextual Bandits (NeuCBs). Using existing results for wide networks, one can readily show a ${\mathcal{O}}(\sqrt{T})$ regret for online regression with square loss, which via the reduction implies a ${\mathcal{O}}(\sqrt{K} T^{3/4})$ regret for NeuCBs. Departing from this standard approach, we first show a $\mathcal{O}(\log T)$ regret for online regression with almost convex losses that satisfy QG (Quadratic Growth) condition, a generalization of the PL (Polyak-\L ojasiewicz) condition, and that have a unique minima. Although not directly applicable to wide networks since they do not have unique minima, we show that adding a suitable small random perturbation to the network predictions surprisingly makes the loss satisfy QG with unique minima. Based on such a perturbed prediction, we show a ${\mathcal{O}}(\log T)$ regret for online regression with both squared loss and KL loss, and subsequently convert these respectively to $\tilde{\mathcal{O}}(\sqrt{KT})$ and $\tilde{\mathcal{O}}(\sqrt{KL^*} + K)$ regret for NeuCB, where $L^*$ is the loss of the best policy. Separately, we also show that existing regret bounds for NeuCBs are $\Omega(T)$ or assume i.i.d. contexts, unlike this work. Finally, our experimental results on various datasets demonstrate that our algorithms, especially the one based on KL loss, persistently outperform existing algorithms.

Graph transformer has been proven as an effective graph learning method for its adoption of attention mechanism that is capable of capturing expressive representations from complex topological and feature information of graphs. Graph transformer conventionally performs dense attention (or global attention) for every pair of nodes to learn node representation vectors, resulting in quadratic computational costs that are unaffordable for large-scale graph data. Therefore, mini-batch training for graph transformers is a promising direction, but limited samples in each mini-batch can not support effective dense attention to encode informative representations. Facing this bottleneck, (1) we start by assigning each node a token list that is sampled by personalized PageRank (PPR) and then apply standard multi-head self-attention only on this list to compute its node representations. This PPR tokenization method decouples model training from complex graph topological information and makes heavy feature engineering offline and independent, such that mini-batch training of graph transformers is possible by loading each node's token list in batches. We further prove this PPR tokenization is viable as a graph convolution network with a fixed polynomial filter and jumping knowledge. However, only using personalized PageRank may limit information carried by a token list, which could not support different graph inductive biases for model training. To this end, (2) we rewire graphs by introducing multiple types of virtual connections through structure- and content-based super nodes that enable PPR tokenization to encode local and global contexts, long-range interaction, and heterophilous information into each node's token list, and then formalize our $\underline{\textbf{V}}$irtual $\underline{\textbf{C}}$onnection $\underline{\textbf{R}}$anking based $\underline{\textbf{Graph}}$ Trans$\underline{\textbf{former}}$ (VCR-Graphormer). Overall, VCR-Graphormer needs $O(m+klogk)$ complexity for graph tokenization as compared to $O(n^{3})$ of previous works. The [code](https://github.com/DongqiFu/VCR-Graphormer) is provided.

Developing a universal model that can effectively harness heterogeneous resources and respond to a wide range of personalized needs has been a longstanding community aspiration. Our daily choices, especially in domains like fashion and retail, are substantially shaped by multi-modal data, such as pictures and textual descriptions. These modalities not only offer intuitive guidance but also cater to personalized user preferences. However, the predominant personalization approaches mainly focus on ID or text-based recommendation problems, failing to comprehend the information spanning various tasks or modalities. In this paper, our goal is to establish a Unified paradigm for Multi-modal Personalization systems (UniMP), which effectively leverages multi-modal data while eliminating the complexities associated with task- and modality-specific customization. We argue that the advancements in foundational generative modeling have provided the flexibility and effectiveness necessary to achieve the objective. In light of this, we develop a generic and extensible personalization generative framework, that can handle a wide range of personalized needs including item recommendation, product search, preference prediction, explanation generation, and further user-guided image generation. Our methodology enhances the capabilities of foundational language models for personalized tasks by seamlessly ingesting interleaved cross-modal user history information, ensuring a more precise and customized experience for users. To train and evaluate the proposed multi-modal personalized tasks, we also introduce a novel and comprehensive benchmark covering a variety of user requirements. Our experiments on the real-world benchmark showcase the model's potential, outperforming competitive methods specialized for each task.

Class imbalance is prevalent in real-world node classification tasks and poses great challenges for graph learning models. Most existing studies are rooted in a class-rebalancing (CR) perspective and address class imbalance with class-wise reweighting or resampling. In this work, we approach the root cause of class-imbalance bias from an topological paradigm. Specifically, we theoretically reveal two **fundamental phenomena in the graph topology** that greatly exacerbate the predictive bias stemming from class imbalance. On this basis, we devise a lightweight topological augmentation framework BAT to mitigate the class-imbalance bias without class rebalancing. Being orthogonal to CR, BAT can function as an **efficient plug-and-play module** that can be seamlessly combined with and significantly boost existing CR techniques. Systematic experiments on real-world imbalanced graph learning tasks show that BAT can deliver up to 46.27% performance gain and up to 72.74% bias reduction over existing techniques. Code, examples, and documentations are available at https://github.com/ZhiningLiu1998/BAT.

Mixup, which generates synthetic training samples on the data manifold, has been shown to be highly effective in augmenting Euclidean data. However, finding a proper data manifold for graph data is non-trivial, as graphs are non-Euclidean data in disparate spaces. Though efforts have been made, most of the existing graph mixup methods neglect the intrinsic geodesic guarantee, thereby generating inconsistent sample-label pairs. To address this issue, we propose GeoMix to mixup graphs on the Gromov-Wasserstein (GW) geodesics. A joint space over input graphs is first defined based on the GW distance, and graphs are then transformed into the GW space through equivalence-preserving transformations. We further show that the linear interpolation of the transformed graph pairs defines a geodesic connecting the original pairs on the GW manifold, hence ensuring the consistency between generated samples and labels. An accelerated mixup algorithm on the approximate low-dimensional GW manifold is further proposed. Extensive experiments show that the proposed GeoMix promotes the generalization and robustness of GNN models.

Heterogeneous data widely exists in various high-impact applications. Domain adaptation and out-of-distribution generalization paradigms have been formulated to handle the data heterogeneity across domains. However, most existing domain adaptation and out-of-distribution generalization algorithms do not explicitly explain how the label information can be adaptively propagated from the source domains to the target domain. Furthermore, little effort has been devoted to theoretically understanding the convergence of existing algorithms based on neural networks.To address these problems, in this paper, we propose a generic distributional network of networks (TENON) framework, where each node of the main network represents an individual domain associated with a domain-specific network. In this case, the edges within the main network indicate the domain similarity, and the edges within each network indicate the sample similarity. The crucial idea of TENON is to characterize the within-domain label smoothness and cross-domain parameter smoothness in a unified framework. The convergence and optimality of TENON are theoretically analyzed. Furthermore, we show that based on the TENON framework, domain adaptation and out-of-distribution generalization can be naturally formulated as transductive and inductive distribution learning problems, respectively. This motivates us to develop two instantiated algorithms (TENON-DA and TENON-OOD) of the proposed TENON framework for domain adaptation and out-of-distribution generalization. The effectiveness and efficiency of TENON-DA and TENON-OOD are verified both theoretically and empirically.

The contextual bandit has been identified as a powerful framework to formulate the recommendation process as a sequential decision-making process, where each item is regarded as an arm and the objective is to minimize the regret of T rounds. In this paper, we study a new problem, Clustering of Neural Bandits, by extending previous work to the arbitrary reward function, to strike a balance between user heterogeneity and user correlations in the recommender system. To solve this problem, we propose a novel algorithm called M-CNB, which utilizes a meta-learner to represent and rapidly adapt to dynamic clusters, along with an informative Upper Confidence Bound (UCB)-based exploration strategy. We provide an instance-dependent performance guarantee for the proposed algorithm that withstands the adversarial context, and we further prove the guarantee is at least as good as state-of-the-art (SOTA) approaches under the same assumptions. In extensive experiments conducted in both recommendation and online classification scenarios, M-CNB outperforms SOTA baselines. This shows the effectiveness of the proposed approach in improving online recommendation and online classification performance.

In the era of big data and Artificial Intelligence, an emerging paradigm is to utilize contrastive self-supervised learning to model large-scale heterogeneous data. Many existing foundation models benefit from the generalization capability of contrastive self-supervised learning by learning compact and high-quality representations without relying on any label information. Amidst the explosive advancements in foundation models across multiple domains, including natural language processing and computer vision, a thorough survey on heterogeneous contrastive learning for the foundation model is urgently needed. In response, this survey critically evaluates the current landscape of heterogeneous contrastive learning for foundation models, highlighting the open challenges and future trends of contrastive learning. In particular, we first present how the recent advanced contrastive learning-based methods deal with view heterogeneity and how contrastive learning is applied to train and fine-tune the multi-view foundation models. Then, we move to contrastive learning methods for task heterogeneity, including pretraining tasks and downstream tasks, and show how different tasks are combined with contrastive learning loss for different purposes. Finally, we conclude this survey by discussing the open challenges and shedding light on the future directions of contrastive learning.

Link prediction is a critical problem in graph learning with broad applications such as recommender systems and knowledge graph completion. Numerous research efforts have been directed at solving this problem, including approaches based on similarity metrics and Graph Neural Networks (GNN). However, most existing solutions are still rooted in conventional supervised learning, which makes it challenging to adapt over time to changing customer interests and to address the inherent dilemma of exploitation versus exploration in link prediction. To tackle these challenges, this paper reformulates link prediction as a sequential decision-making process, where each link prediction interaction occurs sequentially. We propose a novel fusion algorithm, PRB (PageRank Bandits), which is the first to combine contextual bandits with PageRank for collaborative exploitation and exploration. We also introduce a new reward formulation and provide a theoretical performance guarantee for PRB. Finally, we extensively evaluate PRB in both online and offline settings, comparing it with bandit-based and graph-based methods. The empirical success of PRB demonstrates the value of the proposed fusion approach. Our code is released at https://github.com/jiaruzouu/PRB.

Synthesizing time series data is pivotal in modern society, aiding effective decision making and ensuring privacy preservation in various scenarios. Time series are associated with various attributes, including trends, seasonality, and external information such as location. Recent research has predominantly focused on random unconditional synthesis or conditional synthesis. Nonetheless, these paradigms generate time series from scratch and are incapable of manipulating existing time series samples. This paper introduces a novel task, called Time Series Editing (TSE), to synthesize time series by manipulating existing time series. The objective is to modify the given time series according to the specified attributes while preserving other properties unchanged. This task is not trivial due to the inadequacy of data coverage and the intricate relationships between time series and their attributes. To address these issues, we introduce a novel diffusion model, called TEdit. The proposed TEdit is trained using a novel bootstrap learning algorithm that effectively enhances the coverage of the original data. It is also equipped with an innovative multi-resolution modeling and generation paradigm to capture the complex relationships between time series and their attributes. Experimental results demonstrate the efficacy of TEdit for editing specified attributes upon the existing time series data. The project page is at https://seqml.github.io/tse.

Contextual bandit algorithms aim to identify the optimal arm with the highest reward among a set of candidates, based on the accessible contextual information. Among these algorithms, neural contextual bandit methods have shown generally superior performances against linear and kernel ones, due to the representation power of neural networks. However, similar to other neural network applications, neural bandit algorithms can be vulnerable to adversarial attacks or corruptions on the received labels (i.e., arm rewards), which can lead to unexpected performance degradation without proper treatments. As a result, it is necessary to improve the robustness of neural bandit models against potential reward corruptions. In this work, we propose a novel neural contextual bandit algorithm named R-NeuralUCB, which utilizes a novel context-aware Gradient Descent (GD) training strategy to improve the robustness against adversarial reward corruptions. Under over-parameterized neural network settings, we provide regret analysis for R-NeuralUCB to quantify reward corruption impacts, without the commonly adopted arm separateness assumption in existing neural bandit works. We also conduct experiments against baselines on real data sets under different scenarios, in order to demonstrate the effectiveness of our proposed R-NeuralUCB.

_Graph Neural Tangent Kernel_ (GNTK) fuses graph neural networks and graph kernels, simplifies the process of graph representation learning, interprets the training dynamics of graph neural networks, and serves various applications like protein identification, image segmentation, and social network analysis. In practice, graph data carries complex information among entities that inevitably evolves over time, and previous static graph neural tangent kernel methods may be stuck in the sub-optimal solution in terms of both effectiveness and efficiency. As a result, extending the advantage of GNTK to temporal graphs becomes a critical problem. To this end, we propose the temporal graph neural tangent kernel, which not only extends the simplicity and interpretation ability of GNTK to the temporal setting but also leads to rigorous temporal graph classification error bounds. Furthermore, we prove that when the input temporal graph grows over time in the number of nodes, our temporal graph neural tangent kernel will converge in the limit to the _graphon_ NTK value, which implies the transferability and robustness of the proposed kernel method, named **Temp**oral **G**raph **N**eural **T**angent **K**ernel with **G**raphon-**G**uaranteed or **Temp-G$^3$NTK**. In addition to the theoretical analysis, we also perform extensive experiments, not only demonstrating the superiority of Temp-G$^3$NTK in the temporal graph classification task, but also showing that Temp-G^3NTK can achieve very competitive performance in node-level tasks like node classification compared with various SOTA graph kernel and representation learning baselines. Our code is available at https://github.com/kthrn22/TempGNTK.

Knowledge graphs (KGs), which store an extensive number of relational facts (head, relation, tail), serve various applications. While many downstream tasks highly rely on the expressive modeling and predictive embedding of KGs, most of the current KG representation learning methods, where each entity is embedded as a vector in the Euclidean space and each relation is embedded as a transformation, follow an entity ranking protocol. On one hand, such an embedding design cannot capture many-to-many relations. On the other hand, in many retrieval cases, the users wish to get an exact set of answers without any ranking, especially when the results are expected to be precise, e.g., which genes cause an illness. Such scenarios are commonly referred to as "set retrieval". This work presents a pioneering study on the KG set retrieval problem. We show that the set retrieval highly depends on expressive modeling of many-to-many relations, and propose a new KG embedding model SpherE to address this problem. SpherE is based on rotational embedding methods, but each entity is embedded as a sphere instead of a vector. While inheriting the high interpretability of rotational-based models, our SpherE can more expressively model one-to-many, many-to-one, and many-to-many relations. Through extensive experiments, we show that our SpherE can well address the set retrieval problem while still having a good predictive ability to infer missing facts. The code is available at https://github.com/Violet24K/SpherE.

Empirical loss minimization during machine learning training can inadvertently introduce bias, stemming from discrimination and societal prejudices present in the data. To address the shortcomings of traditional fair machine learning methods-which often rely on sensitive information of training data or mandate significant model alterations-we present FairIF, a unique two-stage training framework. Distinctly, FairIF enhances fairness by recalibrating training sample weights using the influence function. Notably, it employs sensitive information from a validation set, rather than the training set, to determine these weights. This approach accommodates situations with missing or inaccessible sensitive training data. Our FairIF ensures fairness across demographic groups by retraining models on the reweighted data. It stands out by offering a plug-and-play solution, obviating the need for changes in model architecture or the loss function. We demonstrate that the fairness performance of FairIF is guaranteed during testing with only a minimal impact on classification performance. Additionally, we analyze that our framework adeptly addresses issues like group size disparities, distribution shifts, and class size discrepancies. Empirical evaluations on three synthetic and five real-world datasets across six model architectures confirm FairIF's efficiency and scalability. The experimental results indicate superior fairness-utility trade-offs compared to other methods, regardless of bias types or architectural variations. Moreover, the adaptability of FairIF to utilize pretrained models for subsequent tasks and its capability to rectify unfairness originating during the pretraining phase are further validated through our experiments.

As data privacy and security attract increasing attention, Federated Recommender System (FRS) offers a solution that strikes a balance between providing high-quality recommendations and preserving user privacy. However, the presence of statistical heterogeneity in FRS, commonly observed due to personalized decision-making patterns, can pose challenges. To address this issue and maximize the benefit of collaborative filtering (CF) in FRS, it is intuitive to consider clustering clients (users) as well as items into different groups and learning group-specific models. Existing methods either resort to client clustering via user representations-risking privacy leakage, or employ classical clustering strategies on item embeddings or gradients, which we found are plagued by the curse of dimensionality. In this paper, we delve into the inefficiencies of the K-Means method in client grouping, attributing failures due to the high dimensionality as well as data sparsity occurring in FRS, and propose CoFedRec, a novel Co-clustering Federated Recommendation mechanism, to address clients heterogeneity and enhance the collaborative filtering within the federated framework. Specifically, the server initially formulates an item membership from the client-provided item networks. Subsequently, clients are grouped regarding a specific item category picked from the item membership during each communication round, resulting in an intelligently aggregated group model. Meanwhile, to comprehensively capture the global inter-relationships among items, we incorporate an additional supervised contrastive learning term based on the server-side generated item membership into the local training phase for each client. Extensive experiments on four datasets are provided, which verify the effectiveness of the proposed CoFedRec.

Effective root cause analysis (RCA) is vital for swiftly restoring services, minimizing losses, and ensuring the smooth operation and management of complex systems. Previous data-driven RCA methods, particularly those employing causal discovery techniques, have primarily focused on constructing dependency or causal graphs for backtracking the root causes. However, these methods often fall short as they rely solely on data from a single modality, thereby resulting in suboptimal solutions. In this work, we propose Mulan, a unified multi-modal causal structure learning method designed to identify root causes in microservice systems. We leverage a log-tailored language model to facilitate log representation learning, converting log sequences into time-series data. To explore intricate relationships across different modalities, we propose a contrastive learning-based approach to extract modality-invariant and modality-specific representations within a shared latent space. Additionally, we introduce a novel key performance indicator-aware attention mechanism for assessing modality reliability and co-learning a final causal graph. Finally, we employ random walk with restart to simulate system fault propagation and identify potential root causes. Extensive experiments on three real-world datasets validate the effectiveness of our proposed method.

Transfer learning refers to the transfer of knowledge or information from a relevant source domain to a target domain. However, most existing transfer learning theories and algorithms focus on IID tasks, where the source/target samples are assumed to be independent and identically distributed. Very little effort is devoted to theoretically studying the knowledge transferability on non-IID tasks, e.g., cross-network mining. To bridge the gap, in this paper, we propose rigorous generalization bounds and algorithms for cross-network transfer learning from a source graph to a target graph. The crucial idea is to characterize the cross-network knowledge transferability from the perspective of the Weisfeiler-Lehman graph isomorphism test. To this end, we propose a novel Graph Subtree Discrepancy to measure the graph distribution shift between source and target graphs. Then the generalization error bounds on cross-network transfer learning, including both cross-network node classification and link prediction tasks, can be derived in terms of the source knowledge and the Graph Subtree Discrepancy across domains. This thereby motivates us to propose a generic graph adaptive network (GRADE) to minimize the distribution shift between source and target graphs for cross-network transfer learning. Experimental results verify the effectiveness and efficiency of our GRADE framework on both cross-network node classification and cross-domain recommendation tasks.

The growth of e-commerce has seen a surge in popularity of platforms like Amazon, eBay, and Taobao. This has given rise to a unique shopping behavior involving baskets - sets of items purchased together. As a less studied interaction mode in the community, the question of how should shopping basket complement personalized recommendation systems remains under-explored. While previous attempts focused on jointly modeling user purchases and baskets, the distinct semantic nature of these elements can introduce noise when directly integrated. This noise negatively impacts the model's performance, further exacerbated by significant noise (e.g., a user is misled to click an item or recognizes it as uninteresting after consuming it) within both user and basket behaviors. In order to cope with the above difficulties, we propose a novel Basket recommendation framework via Noise-tolerated Contrastive Learning, named BNCL, to handle the noise existing in the cross-behavior integration and within-behavior modeling. First, we represent the basket-item interactions as the hypergraph to model the complex basket behavior, where all items appearing in the same basket are treated as a single hyperedge. Second, cross-behavior contrastive learning is designed to suppress the noise during the fusion of diverse behaviors. Next, to further inhibit the within-behavior noise of the user and basket interactions, we propose to exploit invariant properties of the recommenders w.r.t augmentations through within-behavior contrastive learning. A novel consistency-aware augmentation approach is further designed to better identify the noisy interactions with the consideration of the above two types of interactions. Our framework BNCL offers a generic training paradigm that is applicable to different backbones. Extensive experiments on three shopping transaction datasets verify the effectiveness of our proposed method.

In federated learning (FL), multiple clients collaborate to train machine learning models together while keeping their data decentralized. Through utilizing more training data, FL suffers from the potential negative transfer problem: the global FL model may even perform worse than the models trained with local data only. In this paper, we propose FedCollab, a novel FL framework that alleviates negative transfer by clustering clients into non-overlapping coalitions based on their distribution distances and data quantities. As a result, each client only collaborates with the clients having similar data distributions, and tends to collaborate with more clients when it has less data. We evaluate our framework with a variety of datasets, models, and types of non-IIDness. Our results demonstrate that FedCollab effectively mitigates negative transfer across a wide range of FL algorithms and consistently outperforms other clustered FL algorithms.

Fine-tuning a pre-trained language model (PLM) emerges as the predominant strategy in many natural language processing applications. However, even fine-tuning the PLMs and doing inference are expensive, especially on edge devices with low computing power. Some general approaches (e.g. quantization and distillation) have been widely studied to reduce the compute/memory of PLM fine-tuning, while very few one-shot compression techniques are explored. In this paper, we investigate the neural tangent kernel (NTK)–which reveals the gradient descent dynamics of neural networks–of the multilayer perceptrons (MLP) modules in a PLM and propose to coin a lightweight PLM through NTK-approximating MLP fusion. To achieve this, we reconsider the MLP as a bundle of sub-MLPs, and cluster them into a given number of centroids, which can then be restored as a compressed MLP and surprisingly shown to well approximate the NTK of the original PLM. Extensive experiments of PLM fine-tuning on both natural language understanding (NLU) and generation (NLG) tasks are provided to verify the effectiveness of the proposed method MLP fusion. Our code is available at https://github.com/weitianxin/MLP_Fusion.

Recent years have witnessed a surge of research interest in graph machine learning. However, the benchmark datasets available to the field are rather limited in both quantity and diversity, an issue particularly notable given the immense potential applications of graph learning. The lack of diverse benchmark datasets may have biased the development of graph machine learning techniques towards narrow directions. By crowdsourcing novel tasks and datasets, this workshop aims to increase the diversity of graph learning benchmarks, identify new demands of graph machine learning in general, and gain a better synergy of how concrete techniques perform on these benchmarks. Moreover, this workshop offers a platform for discussions of best practices in curating graph learning benchmarks and data-centric approaches for graph learning.

With decentralized data collected from diverse clients, a personalized federated learning paradigm has been proposed for training machine learning models without exchanging raw data from local clients. We dive into personalized federated learning from the perspective of privacy-preserving transfer learning, and identify the limitations of previous personalized federated learning algorithms. First, previous works suffer from negative knowledge transferability for some clients, when focusing more on the overall performance of all clients. Second, high communication costs are required to explicitly learn statistical task relatedness among clients. Third, it is computationally expensive to generalize the learned knowledge from experienced clients to new clients.

Contextual bandits algorithms aim to choose the optimal arm with the highest reward out of a set of candidates based on the contextual information. Various bandit algorithms have been applied to real-world applications due to their ability of tackling the exploitation-exploration dilemma. Motivated by online recommendation scenarios, in this paper, we propose a framework named Graph Neural Bandits (GNB) to leverage the collaborative nature among users empowered by graph neural networks (GNNs). Instead of estimating rigid user clusters as in existing works, we model the "fine-grained" collaborative effects through estimated user graphs in terms of exploitation and exploration respectively. Then, to refine the recommendation strategy, we utilize separate GNN-based models on estimated user graphs for exploitation and adaptive exploration. Theoretical analysis and experimental results on multiple real data sets in comparison with state-of-the-art baselines are provided to demonstrate the effectiveness of our proposed framework.

Deep transfer learning investigates the transfer of knowledge or information from a source domain to a relevant target domain via deep neural networks. In this tutorial, we dive into understanding deep transfer learning from the perspective of knowledge transferability and trustworthiness (e.g., privacy, robustness, fairness, transparency, etc.). To this end, we provide a comprehensive review of state-of-the-art theoretical analysis and algorithms for deep transfer learning. To be specific, we start by introducing the concepts of transferability and trustworthiness in the context of deep transfer learning. Then we summarize recent theories and algorithms for understanding knowledge transferability from two aspects: (1) IID transferability: the samples within each domain are independent and identically distributed (e.g., individual images), and (2) non-IID transferability: The samples within each domain are interdependent (e.g., connected nodes within a graph). In addition to knowledge transferability, we also review the impact of trustworthiness on deep transfer learning, e.g., whether the transferred knowledge is adversarially robust or algorithmically fair, how to transfer the knowledge under privacy-preserving constraints, etc. Finally, we highlight the open questions and future directions for understanding deep transfer learning in real-world applications. We believe this tutorial can benefit researchers and practitioners by rethinking the trade-off between knowledge transferability and trustworthiness in developing trustworthy transfer learning systems.

Personalized federated learning algorithms have shown promising results in adapting models to various distribution shifts. However, most of these methods require labeled data on testing clients for personalization, which is usually unavailable in real-world scenarios. In this paper, we introduce a novel setting called test-time personalized federated learning (TTPFL), where clients locally adapt a global model in an unsupervised way without relying on any labeled data during test-time. While traditional test-time adaptation (TTA) can be used in this scenario, most of them inherently assume training data come from a single domain, while they come from multiple clients (source domains) with different distributions. Overlooking these domain interrelationships can result in suboptimal generalization. Moreover, most TTA algorithms are designed for a specific kind of distribution shift and lack the flexibility to handle multiple kinds of distribution shifts in FL. In this paper, we find that this lack of flexibility partially results from their pre-defining which modules to adapt in the model. To tackle this challenge, we propose a novel algorithm called ATP to adaptively learns the adaptation rates for each module in the model from distribution shifts among source domains. Theoretical analysis proves the strong generalization of ATP. Extensive experiments demonstrate its superiority in handling various distribution shifts including label shift, image corruptions, and domain shift, outperforming existing TTA methods across multiple datasets and model architectures. Our code is available at https://github.com/baowenxuan/ATP.

Meta-learning has been proven an effective learning paradigm for training machine learning models with good generalization ability. Apart from the common practice of uniformly sampling the meta-training tasks, existing methods working on task scheduling strategies are mainly based on pre-defined sampling protocols or the assumed task-model correlations, and greedily make scheduling decisions, which can lead to sub-optimal performance bottlenecks of the meta-model. In this paper, we propose a novel task scheduling framework under Contextual Bandits settings, named BASS, which directly optimizes the task scheduling strategy based on the status of the meta-model. By balancing the exploitation and exploration in meta-learning task scheduling, BASS can help tackle the challenge of limited knowledge about the task distribution during the early stage of meta-training, while simultaneously exploring potential benefits for forthcoming meta-training iterations through an adaptive exploration strategy. Theoretical analysis and extensive experiments are presented to show the effectiveness of our proposed framework.

Transferable graph learning involves knowledge transferability from a source graph to a relevant target graph. The major challenge of transferable graph learning is the distribution shift between source and target graphs induced by individual node attributes and complex graph structures. To solve this problem, in this paper, we propose a generic graph-structured Gaussian process framework (GraphGP) for adaptively transferring knowledge across graphs with either homophily or heterophily assumptions. Specifically, GraphGP is derived from a novel graph structure-aware neural network in the limit on the layer width. The generalization analysis of GraphGP explicitly investigates the connection between knowledge transferability and graph domain similarity. Extensive experiments on several transferable graph learning benchmarks demonstrate the efficacy of GraphGP over state-of-the-art Gaussian process baselines.

In the era of big data, we are often facing the challenge of data heterogeneity and the lack of label information simultaneously. In the financial domain (e.g., fraud detection), the heterogeneous data may include not only numerical data (e.g., total debt and yearly income), but also text and images (e.g., financial statement and invoice images). At the same time, the label information (e.g., fraud transactions) may be missing for building predictive models. To address these challenges, many state-of-the-art multi-view clustering methods have been proposed and achieved outstanding performance. However, these methods typically do not take into consideration the fairness aspect and are likely to generate biased results using sensitive information such as race and gender. Therefore, in this paper, we propose a fairness-aware multi-view clustering method named Fair-MVC. It incorporates the group fairness constraint into the soft membership assignment for each cluster to ensure that the fraction of different groups in each cluster is approximately identical to the entire data set. Meanwhile, we adopt the idea of both contrastive learning and non-contrastive learning and propose novel regularizers to handle heterogeneous data in complex scenarios with missing data or noisy features. Experimental results on real-world data sets demonstrate the effectiveness and efficiency of the proposed framework. We also derive insights regarding the relative performance of the proposed regularizers in various scenarios.

In the big data era, the relationship between entities becomes more complex. Therefore, graph (or network) data attracts increasing research attention for carrying complex relational information. For a myriad of graph mining/learning tasks, graph neural networks (GNNs) have been proven as effective tools for extracting informative node and graph representations, which empowers a broad range of applications such as recommendation, fraud detection, molecule design, and many more. However, real-world scenarios bring pragmatic challenges to GNNs. First, the input graphs are evolving, i.e., the graph structure and node features are time-dependent. Integrating temporal information into the GNNs to enhance their representation power requires additional ingenious designs. Second, the input graphs may be unreliable, noisy, and suboptimal for a variety of downstream graph mining/learning tasks. How could end-users deliberately modify the given graphs (e.g., graph topology and node features) to boost GNNs' utility (e.g., accuracy and robustness)? Inspired by the above two kinds of dynamics, in this tutorial, we focus on topics of natural dynamics and artificial dynamics in GNNs and introduce the related works systematically. After that, we point out some promising but under-explored research problems in the combination of these two dynamics. We hope this tutorial could be beneficial to researchers and practitioners in areas including data mining, machine learning, and general artificial intelligence.

With the widespread development of algorithmic fairness, there has been a surge of research interest that aims to generalize the fairness notions from the attributed data to the relational data (graphs). The vast majority of existing work considers the fairness measure in terms of the low-order connectivity patterns (e.g., edges), while overlooking the higher-order patterns (e.g., k-cliques) and the dynamic nature of real-world graphs. For example, preserving triangles from graph cuts during clustering is the key to detecting compact communities; however, if the clustering algorithm only pays attention to triangle-based compactness, then the returned communities lose the fairness guarantee for each group in the graph. Furthermore, in practice, when the graph (e.g., social networks) topology constantly changes over time, one natural question is how can we ensure the compactness and demographic parity at each timestamp efficiently. To address these problems, we start from the static setting and propose a spectral method that preserves clique connections and incorporates demographic fairness constraints in returned clusters at the same time. To make this static method fit for the dynamic setting, we propose two core techniques, Laplacian Update via Edge Filtering and Searching and Eigen-Pairs Update with Singularity Avoided. Finally, all proposed components are combined into an end-to-end clustering framework named F-SEGA, and we conduct extensive experiments to demonstrate the effectiveness, efficiency, and robustness of F-SEGA.

Personalized PageRank, as a graphical model, has been proven as an effective solution in many applications such as web page search, recommendation, etc. However, in the real world, the setting of personalized PageRank is usually dynamic like the evolving World Wide Web. On the one hand, the outdated PageRank solution can be sub-optimal for ignoring the evolution pattern. On the other hand, solving the solution from the scratch at each timestamp causes costly computation complexity. Hence, in this paper, we aim to solve the Personalized PageRank effectively and efficiently in a fully dynamic setting, i.e., every component in the Personalized PageRank formula is dependent on time. To this end, we propose the EvePPR method that can track the exact personalized PageRank solution at each timestamp in the fully dynamic setting, and we theoretically and empirically prove the accuracy and time complexity of EvePPR. Moreover, we apply EvePPR to solve the dynamic knowledge graph alignment task, where a fully dynamic setting is necessary but complex. The experiments show that EvePPR outperforms the state-of-the-art baselines for similar nodes retrieval across graphs.

Multi-source domain adaptation studies the knowledge transferability from multiple labeled source domains to an unlabeled target domain under a distribution shift. However, little effort has been devoted to studying the adversarial vulnerability of multi-source domain adaptation approaches. Specifically, most existing techniques focus on learning the domain-invariant representation to mitigate the distribution shift across domains. In this paper, we theoretically show that the domain-invariant representation cannot guarantee the success of multi-source domain adaptation, when no labeled samples are available in the target domain. This result motivates us to propose a unified framework (AdaptAttack) for data poisoning adversarial attacks on multi-source domain adaptation. The key idea is to maliciously manipulate the label-informed data distributions of source domains by injecting perceptibly unnoticeable noise into the source data. In addition, it requires that the generated adversarial attacks are invisible to multi-source domain adaptation algorithms, i.e., the source classification errors and marginal discrepancies across domains are not negatively affected. Extensive experiments on public domain adaptation benchmarks confirm the effectiveness and computational efficiency of our proposedAdaptAttackframework in both white-box and black-box attack scenarios.

In the big data era, the relationship between entries becomes more and more complex. Many graph (or network) algorithms have already paid attention to dynamic networks, which are more suitable than static ones for fitting the complex real-world scenarios with evolving structures and features. To contribute to the dynamic network representation learning and mining research, we provide a new bunch of label-adequate, dynamics-meaningful, and attribute-sufficient dynamic networks from the health domain. To be specific, in our proposed repository DPPIN, we totally have 12 individual dynamic network datasets at different scales, and each dataset is a dynamic protein-protein interaction network describing protein-level interactions of yeast cells. We hope these domain-specific node features, structure evolution patterns, and node and graph labels could inspire the regularization techniques to increase the performance of graph machine learning algorithms in a more complex setting. Also, we link potential applications with our DPPIN by designing various dynamic graph experiments, where DPPIN could indicate future research opportunities for some tasks by presenting challenges on state-of-the-art baseline algorithms. Finally, we identify future directions to improve the utility of this repository and welcome constructive inputs from the community. All resources (e.g., data and code) of this work are deployed and publicly available at https://github.com/DongqiFu/DPPIN.

In this paper, we study the dynamic transfer learning problem involving adaptive knowledge transfer from a static source domain to a time evolving target domain. One major challenge is the time evolving relatedness of the source domain and the current target domain as the target domain evolves over time. To address this challenge, we derive a generic error bound on the current target domain with flexible domain discrepancy measures. Moreover, we propose a label-informed $\mathcal{C}$-divergence to measure the shift of joint data distributions (over input features and output labels) across domains. The resulting tighter error bound with $\mathcal{C}$-divergence motivates us to develop a novel dynamic transfer learning algorithm TransLATE. Empirical results on various data sets confirm the effectiveness of our proposed algorithm in modeling the time evolving target domain.

Disinformation refers to false information deliberately spread to influence the general public, and the negative impact of disinformation on society can be observed in numerous issues, such as political agendas and manipulating financial markets. In this paper, we identify prevalent challenges and advances related to automated disinformation detection from multiple aspects and propose a comprehensive and explainable disinformation detection framework called DISCO. It leverages the heterogeneity of disinformation and addresses the opaqueness of prediction. Then we provide a demonstration of DISCO on a real-world fake news detection task with satisfactory detection accuracy and explanation. The demo video and source code of DISCO is now publicly available https://github.com/DongqiFu/DISCO. We expect that our demo could pave the way for addressing the limitations of identification, comprehension, and explainability as a whole.

Learning on graphs (LOG) plays a pivotal role in various high-impact application domains. The past decades have developed tremendous theories, algorithms, and open-source systems in answering what/who questions on graphs. However, recent studies reveal that the state-of-the-art techniques for learning on graphs (LOG) are often not trustworthy in practice with respect to several social aspects (e.g., fairness, transparency, security). A natural research question to ask is: how can we make learning algorithms on graphs trustworthy? To answer this question, we propose a paradigm shift, from answering what and who LOG questions to understanding how and why LOG questions. The TrustLOG workshop provides a venue for presenting, discussing, and promoting frontier research on trustworthy learning on graphs. Moreover, TrustLOG will serve as an impulse for the LOG community to identify novel research problems and shed new light on future directions.

Federated learning learns a neural network model by aggregating the knowledge from a group of distributed clients under the privacy-preserving constraint. In this work, we show that this paradigm might inherit the adversarial vulnerability of the centralized neural network, i.e., it has deteriorated performance on adversarial examples when the model is deployed. This is even more alarming when federated learning paradigm is designed to approximate the updating behavior of a centralized neural network. To solve this problem, we propose an adversarially robust federated learning framework, named Fed_BVA, with improved server and client update mechanisms. This is motivated by our observation that the generalization error in federated learning can be naturally decomposed into the bias and variance triggered by multiple clients' predictions. Thus, we propose to generate the adversarial examples via maximizing the bias and variance during server update, and learn the adversarially robust model updates with those examples during client update. As a result, an adversarially robust neural network can be aggregated from these improved local clients' model updates. The experiments are conducted on multiple benchmark data sets using several prevalent neural network models, and the empirical results show that our framework is robust against white-box and black-box adversarial corruptions under both IID and non-IID settings.

Graph pre-training strategies have been attracting a surge of attention in the graph mining community, due to their flexibility in parameterizing graph neural networks (GNNs) without any label information. The key idea lies in encoding valuable information into the backbone GNNs, by predicting the masked graph signals extracted from the input graphs. In order to balance the importance of diverse graph signals (e.g., nodes, edges, subgraphs), the existing approaches are mostly hand-engineered by introducing hyperparameters to re-weight the importance of graph signals. However, human interventions with sub-optimal hyperparameters often inject additional bias and deteriorate the generalization performance in the downstream applications. This paper addresses these limitations from a new perspective, i.e., deriving curriculum for pre-training GNNs. We propose an end-to-end model named MentorGNN that aims to supervise the pre-training process of GNNs across graphs with diverse structures and disparate feature spaces. To comprehend heterogeneous graph signals at different granularities, we propose a curriculum learning paradigm that automatically re-weighs graph signals in order to ensure a good generalization in the target domain. Moreover, we shed new light on the problem of domain adaption on relational data (i.e., graphs) by deriving a natural and interpretable upper bound on the generalization error of the pre-trained GNNs. Extensive experiments on a wealth of real graphs validate and verify the performance of MentorGNN.

Transfer learning refers to the transfer of knowledge or information from a relevant source task to a target task. However, most existing works assume both tasks are sampled from a stationary task distribution, thereby leading to the sub-optimal performance for dynamic tasks drawn from a non-stationary task distribution in real scenarios. To bridge this gap, in this paper, we study a more realistic and challenging transfer learning setting with dynamic tasks, i.e., source and target tasks are continuously evolving over time. We theoretically show that the expected error on the dynamic target task can be tightly bounded in terms of source knowledge and consecutive distribution discrepancy across tasks. This result motivates us to propose a generic meta-learning framework L2E for modeling the knowledge transferability on dynamic tasks. It is centered around a task-guided meta-learning problem with a group of meta-pairs of tasks, based on which we are able to learn the prior model initialization for fast adaptation on the newest target task. L2E enjoys the following properties: (1) effective knowledge transferability across dynamic tasks; (2) fast adaptation to the new target task; (3) mitigation of catastrophic forgetting on historical target tasks; and (4) flexibility in incorporating any existing static transfer learning algorithms. Extensive experiments on various image data sets demonstrate the effectiveness of the proposed L2E framework.

Open-set domain adaptation aims to improve the generalization performance of a learning algorithm on a target task of interest by leveraging the label information from a relevant source task with only a subset of classes. However, most existing works are designed for the static setting, and can be hardly extended to the dynamic setting commonly seen in many real-world applications. In this paper, we focus on the more realistic open-set domain adaptation setting with a static source task and a time evolving target task where novel unknown target classes appear over time. Specifically, we show that the classification error of the new target task can be tightly bounded in terms of positive-unlabeled classification errors for historical tasks and open-set domain discrepancy across tasks. By empirically minimizing the upper bound of the target error, we propose a novel positive-unlabeled learning based algorithm named OuterAdapter for dynamic open-set domain adaptation with time evolving unknown classes. Extensive experiments on various data sets demonstrate the effectiveness and efficiency of our proposed OuterAdapter algorithm over state-of-the-art domain adaptation baselines.

Graph metric learning methods aim to learn the distance metric over graphs such that similar (e.g., same class) graphs are closer and dissimilar (e.g., different class) graphs are farther apart. This is of critical importance in many graph classification applications such as drug discovery and epidemics categorization. Most, if not all, graph metric learning techniques consider the input graph as static, and largely ignore the intrinsic dynamics of temporal graphs. However, in practice, a graph typically has heterogeneous dynamics (e.g., microscopic and macroscopic evolution patterns). As such, labeling a temporal graph is usually expensive and also requires background knowledge. To learn a good metric over temporal graphs, we propose a temporal graph metric learning framework, Temp-GFSM. With only a few labeled temporal graphs, Temp-GFSM outputs a good metric that can accurately classify different temporal graphs and be adapted to discover new subspaces for unseen classes. Each proposed component in Temp-GFSM answers the following questions: What patterns are evolving in a temporal graph? How to weigh these patterns to represent the characteristics of different temporal classes? And how to learn the metric with the guidance from only a few labels? Finally, the experimental results on real-world temporal graph classification tasks from various domains show the effectiveness of our Temp-GFSM.

Contextual bandits aim to identify among a set of arms the optimal one with the highest reward based on their contextual information. Motivated by the fact that the arms usually exhibit group behaviors and the mutual impacts exist among groups, we introduce a new model, Arm Group Graph (AGG), where the nodes represent the groups of arms and the weighted edges formulate the correlations among groups. To leverage the rich information in AGG, we propose a bandit algorithm, AGG-UCB, where the neural networks are designed to estimate rewards, and we propose to utilize graph neural networks (GNN) to learn the representations of arm groups with correlations. To solve the exploitation-exploration dilemma in bandits, we derive a new upper confidence bound (UCB) built on neural networks (exploitation) for exploration. Furthermore, we prove that AGG-UCB can achieve a near-optimal regret bound with over-parameterized neural networks, and provide the convergence analysis of GNN with fully-connected layers which may be of independent interest. In the end, we conduct extensive experiments against state-of-the-art baselines on multiple public data sets, showing the effectiveness of the proposed algorithm.

In recommender systems, one common challenge is the cold-start problem, where interactions are very limited for fresh users in the systems. To address this challenge, recently, many works introduce the meta-optimization idea into the recommendation scenarios, i.e. learning to learn the user preference by only a few past interaction items. The core idea is to learn global shared meta-initialization parameters for all users and rapidly adapt them into local parameters for each user respectively. They aim at deriving general knowledge across preference learning of various users, so as to rapidly adapt to the future new user with the learned prior and a small amount of training data. However, previous works have shown that recommender systems are generally vulnerable to bias and unfairness. Despite the success of meta-learning at improving the recommendation performance with cold-start, the fairness issues are largely overlooked.

With the advent of big data across multiple high-impact applications, we are often facing the challenge of complex heterogeneity. The newly collected data usually consist of multiple modalities and are characterized with multiple labels, thus exhibiting the co-existence of multiple types of heterogeneity. Although state-of-the-art techniques are good at modeling the complex heterogeneity with sufficient label information, such label information can be quite expensive to obtain in real applications. Recently, researchers pay great attention to contrastive learning due to its prominent performance by utilizing rich unlabeled data. However, existing work on contrastive learning is not able to address the problem of false-negative pairs, i.e., some 'negative' pairs may have similar representations if they have the same label. To overcome the issues, in this paper, we propose a unified heterogeneous learning framework, which combines both the weighted unsupervised contrastive loss and the weighted supervised contrastive loss to model multiple types of heterogeneity. We first provide a theoretical analysis showing that the vanilla contrastive learning loss easily leads to the sub-optimal solution in the presence of false-negative pairs, whereas the proposed weighted loss could automatically adjust the weight based on the similarity of the learned representations to mitigate this issue. Experimental results on real-world data sets demonstrate the effectiveness and the efficiency of the proposed framework modeling multiple types of heterogeneity.

In this paper, we study the problem of domain adaptation regression, which learns a regressor for a target domain by leveraging the knowledge from a relevant source domain. We start by proposing a distribution-informed neural network, which aims to build distribution-aware relationship of inputs and outputs from different domains. This allows us to develop a simple domain adaptation regression framework, which subsumes popular domain adaptation approaches based on domain invariant representation learning, reweighting, and adaptive Gaussian process. The resulting findings not only explain the connections of existing domain adaptation approaches, but also motivate the efficient training of domain adaptation approaches with overparameterized neural networks. We also analyze the convergence and generalization error bound of our framework based on the distribution-informed neural network. Specifically, our generalization bound focuses explicitly on the maximum mean discrepancy in the RKHS induced by the neural tangent kernel of distribution-informed neural network. This is in sharp contrast to the existing work which relies on domain discrepancy in the latent feature space heuristically formed by one or several hidden neural layers. The efficacy of our framework is also empirically verified on a variety of domain adaptation regression benchmarks.

We improve the theoretical and empirical performance of neural-network(NN)-based active learning algorithms for the non-parametric streaming setting. In particular, we introduce two regret metrics by minimizing the population loss that are more suitable in active learning than the one used in state-of-the-art (SOTA) related work. Then, the proposed algorithm leverages the powerful representation of NNs for both exploitation and exploration, has the query decision-maker tailored for $k$-class classification problems with the performance guarantee, utilizes the full feedback, and updates parameters in a more practical and efficient manner. These careful designs lead to an instance-dependent regret upper bound, roughly improving by a multiplicative factor $O(\log T)$ and removing the curse of input dimensionality. Furthermore, we show that the algorithm can achieve the same performance as the Bayes-optimal classifier in the long run under the hard-margin setting in classification problems. In the end, we use extensive experiments to evaluate the proposed algorithm and SOTA baselines, to show the improved empirical performance.

Active learning theories and methods have been extensively studied in classical statistical learning settings. However, deep active learning, i.e., active learning with deep learning models, is usually based on empirical criteria without solid theoretical justification, thus suffering from heavy doubts when some of those fail to provide benefits in applications. In this paper, by exploring the connection between the generalization performance and the training dynamics, we propose a theory-driven deep active learning method (dynamicAL) which selects samples to maximize training dynamics. In particular, we prove that the convergence speed of training and the generalization performance is positively correlated under the ultra-wide condition and show that maximizing the training dynamics leads to a better generalization performance. Furthermore, to scale up to large deep neural networks and data sets, we introduce two relaxations for the subset selection problem and reduce the time complexity from polynomial to constant. Empirical results show that dynamicAL not only outperforms the other baselines consistently but also scales well on large deep learning models. We hope our work inspires more attempts in bridging the theoretical findings of deep networks and practical impacts in deep active learning applications.

This paper targets at improving the generalizability of hypergraph neural networks in the low-label regime, through applying the contrastive learning approach from images/graphs (we refer to it as HyperGCL). We focus on the following question: How to construct contrastive views for hypergraphs via augmentations? We provide the solutions in two folds. First, guided by domain knowledge, we fabricate two schemes to augment hyperedges with higher-order relations encoded, and adopt three vertex augmentation strategies from graph-structured data. Second, in search of more effective views in a data-driven manner, we for the first time propose a hypergraph generative model to generate augmented views, and then an end-to-end differentiable pipeline to jointly learn hypergraph augmentations and model parameters. Our technical innovations are reflected in designing both fabricated and generative augmentations of hypergraphs. The experimental findings include: (i) Among fabricated augmentations in HyperGCL, augmenting hyperedges provides the most numerical gains, implying that higher-order information in structures is usually more downstream-relevant; (ii) Generative augmentations do better in preserving higher-order information to further benefit generalizability; (iii) HyperGCL also boosts robustness and fairness in hypergraph representation learning. Codes are released at https://github.com/weitianxin/HyperGCL.

With the pervasive use of information technologies, the increasing availability of data provides new opportunities for understanding user behaviors. Unearthing anomalies in user behavior is of particular importance as it helps signal harmful incidents such as network intrusions, terrorist activities, and financial frauds. In this article, we survey state-of-the-art research work in visual analytics of anomalous user behaviors and classify them into four application domains, which are social interaction, travel, network communication, and financial transaction. We further examine the research work in each category in terms of data types, visualization techniques, and interactive analysis methods. We hope that our survey can provide systematic guidelines for researchers and practitioners to find effective solutions to their research problems in specific application domains. Finally, we discuss trends of academic interest over the past decades and suggest potential directions across visual analytics of these user behaviors for future research.

For time series data, certain types of outliers are intrinsically more harmful for parameter estimation and future predictions than others, irrespective of their frequency. In this paper, for the first time, we study the characteristics of such outliers through the lens of the influence functional from robust statistics. In particular, we consider the input time series as a contaminated process, with the recurring outliers generated from an unknown contaminating process. Then we leverage the influence functional to understand the impact of the contaminating process on parameter estimation. The influence functional results in a multi-dimensional vector that measures the sensitivity of the predictive model to the contaminating process, which can be challenging to interpret especially for models with a large number of parameters. To this end, we further propose a comprehensive single-valued metric (the SIF) to measure outlier impacts on future predictions. It provides a quantitative measure regarding the outlier impacts, which can be used in a variety of scenarios, such as the evaluation of outlier detection methods, the creation of more harmful outliers, etc. The empirical results on multiple real data sets demonstrate the effectivenss of the proposed SIF metric.

Contextual multi-armed bandit has shown to be an effective tool in recommender systems. In this paper, we study a novel problem of multi-facet bandits involving a group of bandits, each characterizing the users' needs from one unique aspect. In each round, for the given user, we need to select one arm from each bandit, such that the combination of all arms maximizes the final reward. This problem can find immediate applications in E-commerce, healthcare, etc. To address this problem, we propose a novel algorithm, named MuFasa, which utilizes an assembled neural network to jointly learn the underlying reward functions of multiple bandits. It estimates an Upper Confidence Bound (UCB) linked with the expected reward to balance between exploitation and exploration. Under mild assumptions, we provide the regret analysis of MuFasa. It can achieve the near-optimal Õ((K + 1) √T) regret bound where K is the number of bandits and T is the number of played rounds. Furthermore, we conduct extensive experiments to show that MuFasa outperforms strong baselines on real-world data sets.

Unsupervised domain adaptation has been successfully applied across multiple high-impact applications, since it improves the generalization performance of a learning algorithm when the source and target domains are related. However, the adversarial vulnerability of domain adaptation models has largely been neglected. Most existing unsupervised domain adaptation algorithms might be easily fooled by an adversary, resulting in deteriorated prediction performance on the target domain, when transferring the knowledge from a maliciously manipulated source domain.

Recommender systems are powerful tools for information filtering with the ever-growing amount of online data. Despite its success and wide adoption in various web applications and personalized products, many existing recommender systems still suffer from multiple drawbacks such as large amount of unobserved feedback, poor model convergence, etc. These drawbacks of existing work are mainly due to the following two reasons: first, the widely used negative sampling strategy, which treats the unlabeled entries as negative samples, is invalid in real-world settings; second, all training samples are retrieved from the discrete observations, and the underlying true distribution of the users and items is not learned.

Graph Neural Networks (GNNs) have achieved state-of-the-art performance in many high-impact applications such as fraud detection, information retrieval, and recommender systems due to their powerful representation learning capabilities. Some nascent efforts have been concentrated on simplifying the structures of GNN models, in order to reduce the computational complexity. However, the dynamic nature of these applications requires GNN structures to be evolving over time, which has been largely overlooked so far. To bridge this gap, in this paper, we propose a simplified and dynamic graph neural network model, called SDG. It is efficient, effective, and provides interpretable predictions. In particular, in SDG, we replace the traditional message-passing mechanism of GNNs with the designed dynamic propagation scheme based on the personalized PageRank tracking process. We conduct extensive experiments and ablation studies to demonstrate the effectiveness and efficiency of our proposed SDG. We also design a case study on fake news detection to show the interpretability of SDG.

We study identifying user clusters in contextual multi-armed bandits (MAB). Contextual MAB is an effective tool for many real applications, such as content recommendation and online advertisement. In practice, user dependency plays an essential role in the user’s actions, and thus the rewards. Clustering similar users can improve the quality of reward estimation, which in turn leads to more effective content recommendation and targeted advertising. Different from traditional clustering settings, we cluster users based on the unknown bandit parameters, which will be estimated incrementally. In particular, we define the problem of cluster detection in contextual MAB, and propose a bandit algorithm, LOCB, embedded with local clustering procedure. And, we provide theoretical analysis about LOCB in terms of the correctness and efficiency of clustering and its regret bound. Finally, we evaluate the proposed algorithm from various aspects, which outperforms state-of-the-art baselines.

In this paper, we identify and study an important problem of gradient item retrieval. We define the problem as retrieving a sequence of items with a gradual change on a certain attribute, given a reference item and a modification text. For example, after a customer saw a white dress, she/he wants to buy a similar one but more floral on it. The extent of ”more floral” is subjective, thus prompting one floral dress is hard to satisfy the customer’s needs. A better way is to present a sequence of products with increasingly floral attributes based on the white dress, and allow the customer to select the most satisfactory one from the sequence. Existing item retrieval methods mainly focus on whether the target items appear at the top of the retrieved sequence, but ignore the demand for retrieving a sequence of products with gradual change on a certain attribute. To deal with this problem, we propose a weakly-supervised method that can learn a disentangled item representation from user-item interaction data and ground the semantic meaning of attributes to dimensions of the item representation. Our method takes a reference item and a modification as a query. During inference, we start from the reference item and ”walk” along the direction of the modification in the item representation space to retrieve a sequence of items in a gradient manner. We demonstrate our proposed method can achieve disentanglement through weak supervision. Besides, we empirically show that an item sequence retrieved by our method is gradually changed on an indicated attribute and, in the item retrieval task, our method outperforms existing approaches on three different datasets.

Many real-world applications involve data from multiple modalities and thus exhibit the view heterogeneity. For example, user modeling on social media might leverage both the topology of the underlying social network and the content of the users’ posts; in the medical domain, multiple views could be X-ray images taken at different poses. To date, various techniques have been proposed to achieve promising results, such as canonical correlation analysis based methods, etc. In the meanwhile, it is critical for decision-makers to be able to understand the prediction results from these methods. For example, given the diagnostic result that a model provided based on the X-ray images of a patient at different poses, the doctor needs to know why the model made such a prediction. However, state-of-the-art techniques usually suffer from the inability to utilize the complementary information of each view and to explain the predictions in an interpretable manner.

Modeling and exploring high-order connectivity patterns, also called network motifs, are essential for understanding the fundamental structures that control and mediate the behavior of many complex systems. For example, in social networks, triangles have been proven to play the fundamental role in understanding social network communities; in online transaction networks, detecting directed looped transactions helps identify money laundering activities; in personally identifiable information networks, the star-shaped structures may correspond to a set of synthetic identities. Despite the ubiquity of such high-order structures, many existing graph clustering methods are either not designed for the high-order connectivity patterns, or suffer from the prohibitive computational cost when modeling high-order structures in the large-scale networks. This article generalizes the challenges in multiple dimensions. First ( Model ), we introduce the notion of high-order conductance, and define the high-order diffusion core, which is based on a high-order random walk induced by the user-specified high-order network structure. Second ( Algorithm ), we propose a novel high-order structure-preserving graph clustering framework named HOSGRAP , which partitions the graph into structure-rich clusters in polylogarithmic time with respect to the number of edges in the graph. Third ( Generalization ), we generalize our proposed algorithm to solve the real-world problems on various types of graphs, such as signed graphs, bipartite graphs, and multi-partite graphs. Experimental results on both synthetic and real graphs demonstrate the effectiveness and efficiency of the proposed algorithms.

Many statistical learning models hold an assumption that the training data and the future unlabeled data are drawn from the same distribution. However, this assumption is difficult to fulfill in real-world scenarios and creates barriers in reusing existing labels from similar application domains. Transfer Learning is intended to relax this assumption by modeling relationships between domains, and is often applied in deep learning applications to reduce the demand for labeled data and training time. Despite recent advances in exploring deep learning models with visual analytics tools, little work has explored the issue of explaining and diagnosing the knowledge transfer process between deep learning models. In this paper, we present a visual analytics framework for the multi-level exploration of the transfer learning processes when training deep neural networks. Our framework establishes a multi-aspect design to explain how the learned knowledge from the existing model is transferred into the new learning task when training deep neural networks. Based on a comprehensive requirement and task analysis, we employ descriptive visualization with performance measures and detailed inspections of model behaviors from the statistical, instance, feature, and model structure levels. We demonstrate our framework through two case studies on image classification by fine-tuning AlexNets to illustrate how analysts can utilize our framework.

Encoding a large-scale network into a low-dimensional space is a fundamental step for various network analytic problems, such as node classification, link prediction, community detection, etc. Existing methods focus on learning the network representation from either the static graphs or time-aggregated graphs (e.g., time-evolving graphs). However, many real systems are not static or time-aggregated as the nodes and edges are timestamped and dynamically changing over time. For examples, in anti-money laundering analysis, cycles formed with time-ordered transactions might be red flags in online transaction networks; in novelty detection, a star-shaped structure appearing in a short burst might be an underlying hot topic in social networks. Existing embedding models might not be able to well preserve such fine-grained network dynamics due to the incapability of dealing with continuous-time and the negligence of fine-grained interactions. To bridge this gap, in this paper, we propose a fine-grained temporal network embedding framework named FiGTNE, which aims to learn a comprehensive network representation that preserves the rich and complex network context in the temporal network. In particular, we start from the notion of fine-grained temporal networks, where the temporal network can be represented as a series of timestamped nodes and edges. Then, we propose the time-reinforced random walk (TRRW) with a bi-level context sampling strategy to explore the essential structures and temporal contexts in temporal networks. Extensive experimental results on real graphs demonstrate the efficacy of our FiGTNE framework.

Networks are natural representations in modeling adversarial activities, such as smuggling, human trafficking, and illegal arms dealing. However, such activities are often covert and embedded across multiple domains and sources. They are generally not detectable and recognizable from the perspective of an isolated network, and only become apparent when multiple networks are analyzed in a unified m anner. T o t his e nd, we propose Complex Analytics of Network of Networks (CANON), a mathematical and computational framework for modeling adversarial activities from large-scale, multi-sourced data inputs. Central to our framework is a network-of-networks model, where nodes and edges can be defined across different domains and at multiple resolutions. Based on this model, we address the key challenges in modeling adversarial activities via four technical components, including optimization-based network alignment, network embedding and conditioning, approximate subgraph matching, and investigative subgraph discovery.In this paper, we describe the design and implementation of the individual components as well as integrating these components into a unified system using a modular microservice architecture. Extensive experiments have been conducted in both synthetics and real-world datasets to demonstrate the effectiveness of our proposed system under the DARPA Modeling Adversarial Activity (MAA) program.

Network embedding has demonstrated effective empirical performance for various network mining tasks such as node classification, link prediction, clustering, and anomaly detection. However, most of these algorithms focus on the single-view network scenario. From a real-world perspective, one individual node can have different connectivity patterns in different networks. For example, one user can have different relationships on Twitter, Facebook, and LinkedIn due to varying user behaviors on different platforms. In this case, jointly considering the structural information from multiple platforms (i.e., multiple views) can potentially lead to more comprehensive node representations, and eliminate noises and bias from a single view. In this paper, we propose a view-adversarial framework to generate comprehensive and robust multi-view network representations named VANE, which is based on two adversarial games. The first adversarial game enhances the comprehensiveness of the node representation by discriminating the view information which is obtained from the subgraph induced by neighbors of that node. The second adversarial game improves the robustness of the node representation with the challenging of fake node representations from the generative adversarial net. We conduct extensive experiments on downstream tasks with real-world multi-view networks, which shows that our proposed VANE framework significantly outperforms other baseline methods.

In this paper, we study the problem of outlier arm detection in multi-armed bandit settings, which finds plenty of applications in many high-impact domains such as finance, healthcare, and online advertising. For this problem, a learner aims to identify the arms whose expected rewards deviate significantly from most of the other arms. Different from existing work, we target the generic outlier arms or outlier arm groups whose expected rewards can be larger, smaller, or even in between those of normal arms. To this end, we start by providing a comprehensive definition of such generic outlier arms and outlier arm groups. Then we propose a novel pulling algorithm named GOLD to identify such generic outlier arms. It builds a real-time neighborhood graph based on upper confidence bounds and catches the behavior pattern of outliers from normal arms. We also analyze its performance from various aspects. In the experiments conducted on both synthetic and real-world data sets, the proposed algorithm achieves 98% accuracy while saving 83% exploration cost on average compared with state-of-the-art techniques.

Graph motifs are subgraph patterns that occur in complex networks, which are of key importance for gaining deep insights into the structure and functionality of the graph. Motif clustering aims at finding clusters consisting of dense motif patterns. It is commonly used in various application domains, ranging from social networks to collaboration networks, from market-basket analysis to neuroscience applications. More recently, local clustering techniques have been proposed for motif-aware clustering, which focuses on a small neighborhood of the input seed node instead of the entire graph. However, most of these techniques are designed for static graphs and may render sub-optimal results when applied to large time-evolving graphs. To bridge this gap, in this paper, we propose a novel framework, Local Motif Clustering on Time-Evolving Graphs (L-MEGA), which provides the evolution pattern of the local motif cluster in an effective and efficient way. The core of L-MEGA is approximately tracking the temporal evolution of the local motif cluster via novel techniques such as edge filtering, motif push operation, and incremental sweep cut. Furthermore, we theoretically analyze the efficiency and effectiveness of these techniques on time-evolving graphs. Finally, we evaluate the L-MEGA framework via extensive experiments on both synthetic and real-world temporal networks.

Algorithmic bias and fairness in the context of graph mining have largely remained nascent. The sparse literature on fair graph mining has almost exclusively focused on group-based fairness notation. However, the notion of individual fairness, which promises the fairness notion at a much finer granularity, has not been well studied. This paper presents the first principled study of Individual Fairness on gRaph Mining (InFoRM). First, we present a generic definition of individual fairness for graph mining which naturally leads to a quantitative measure of the potential bias in graph mining results. Second, we propose three mutually complementary algorithmic frameworks to mitigate the proposed individual bias measure, namely debiasing the input graph, debiasing the mining model and debiasing the mining results. Each algorithmic framework is formulated from the optimization perspective, using effective and efficient solvers, which are applicable to multiple graph mining tasks. Third, accommodating individual fairness is likely to change the original graph mining results without the fairness consideration. We conduct a thorough analysis to develop an upper bound to characterize the cost (i.e., the difference between the graph mining results with and without the fairness consideration). We perform extensive experimental evaluations on real-world datasets to demonstrate the efficacy and generality of the proposed methods.

Deep graph generative models have recently received a surge of attention due to its superiority of modeling realistic graphs in a variety of domains, including biology, chemistry, and social science. Despite the initial success, most, if not all, of the existing works are designed for static networks. Nonetheless, many realistic networks are intrinsically dynamic and presented as a collection of system logs (i.e., timestamped interactions/edges between entities), which pose a new research direction for us: how can we synthesize realistic dynamic networks by directly learning from the system logs? In addition, how can we ensure the generated graphs preserve both the structural and temporal characteristics of the real data?

The need for annotated labels to train machine learning models led to a surge in crowdsourcing - collecting labels from non-experts. Instead of annotating from scratch, given an imperfect labeled set, how can we leverage the label information obtained from amateur crowd workers to improve the data quality? Furthermore, is there a way to teach the amateur crowd workers using this imperfect labeled set in order to improve their labeling performance? In this paper, we aim to answer both questions via a novel interactive teaching framework, which uses visual explanations to simultaneously teach and gauge the confidence level of the crowd workers.

Financial time series analysis plays a central role in optimizing investment decision and hedging market risks. This is a challenging task as the problems are always accompanied by dual-level (i.e, data-level and task-level) heterogeneity. For instance, in stock price forecasting, a successful portfolio with bounded risks usually consists of a large number of stocks from diverse domains (e.g, utility, information technology, healthcare, etc.), and forecasting stocks in each domain can be treated as one task; within a portfolio, each stock is characterized by temporal data collected from multiple modalities (e.g, finance, weather, and news), which corresponds to the data-level heterogeneity. Furthermore, the finance industry follows highly regulated processes, which require prediction models to be interpretable, and the output results to meet compliance. Therefore, a natural research question is how to build a model that can achieve satisfactory performance on such multi-modality multi-task learning problems, while being able to provide comprehensive explanations for the end users.

A dynamic network refers to a graph structure whose nodes and/or links dynamically change over time. Existing visualization and analysis techniques focus mainly on summarizing and revealing the primary evolution patterns of the network structure. Little work focuses on detecting anomalous changing patterns in the dynamic network, the rare occurrence of which could damage the development of the entire structure. In this study, we introduce the first visual analysis system RCAnalyzer designed for detecting rare changes of sub-structures in a dynamic network. The proposed system employs a rare category detection algorithm to identify anomalous changing structures and visualize them in the context to help oracles examine the analysis results and label the data. In particular, a novel visualization is introduced, which represents the snapshots of a dynamic network in a series of connected triangular matrices. Hierarchical clustering and optimal tree cut are performed on each matrix to illustrate the detected rare change of nodes and links in the context of their surrounding structures. We evaluate our technique via a case study and a user study. The evaluation results verify the effectiveness of our system.

Collective matrix completion refers to the problem of simultaneously predicting the missing entries in multiple matrices by leveraging the cross-matrix information. It finds abundant applications in various domains such as recommender system, dimensionality reduction, and image recovery. Most of the existing work represents the cross-matrix information in a shared latent structure constrained by the Euclidean-based pairwise similarity, which may fail to capture the nonlinear relationship of the data. To address this problem, in this paper, we propose a new collective matrix completion framework, named C4, which uses the graph spectral filters to capture the non-Euclidean cross-matrix information. To the best of our knowledge, this is the first effort to represent the cross-matrix information in the graph spectral domain. We benchmark our model against 8 recent models on 10 real-world data sets, and our model outperforms state-of-the-art methods in most tasks.

Many complex systems with relational data can be naturally represented as dynamic processes on graphs, with the addition/deletion of nodes and edges over time. For such graphs, network embedding provides an important class of tools for leveraging the node proximity to learn a low-dimensional representation before using the off-the-shelf machine learning models. However, for dynamic graphs, most, if not all, embedding approaches rely on various hyper-parameters to extract spatial and temporal context information, which differ from task to task and from data to data. Besides, many regulated industries (e.g., finance, health care) require the learning models to be interpretable and the output results to meet compliance. Therefore, a natural research question is how we can jointly model the spatial and temporal context information and learn a unique network representation, while being able to provide interpretable inference over the observed data. To address this question, we propose a generic graph attention neural mechanism named STANE, which guides the context sampling process to focus on the crucial part of the data. Moreover, to interpret the network embedding results, STANE enables the end users to investigate the graph context distributions along three dimensions (i.e., nodes, training window length, and time). We perform extensive experiments regarding quantitative evaluation and case studies, which demonstrate the effectiveness and interpretability of STANE.

Networks are ubiquitous in many real-world applications due to their capability of representing the rich information in the data. One fundamental problem of network analysis is to learn a low- dimensional vector representation for nodes within the attributed networks. However, there is little work theoretically considering the information heterogeneity from the attributed networks, and most of the existing attributed network embedding techniques are able to capture at most k-th order node proximity, thus leading to the information loss of the long-range spatial dependencies between individual nodes across the entire network. To address the above problems, in this paper, we propose a novel MAnifold-RegularIzed Network Embedding (MARINE) algorithm inspired by minimizing the information discrepancy in a Reproducing Kernel Hilbert Space via Maximum Mean Discrepancy. In particular, we show that MARINE recursively aggregates the graph structure information as well as individual node attributes from the entire network, and thereby preserves the long-range spatial dependencies between nodes across the network. The experimental results on real networks demonstrate the effectiveness and efficiency of the proposed MARINE algorithm over state-of-the-art embedding methods.

In this paper, we propose a deep multi-Task learning model based on Adversarial-and-COoperative nets (TACO). The goal is to use an adversarial-and-cooperative strategy to decouple the task-common and task-specific knowledge, facilitating the fine-grained knowledge sharing among tasks. TACO accommodates multiple game players, i.e., feature extractors, domain discriminator, and tri-classifiers. They play the MinMax games adversarially and cooperatively to distill the task-common and task-specific features, while respecting their discriminative structures. Moreover, it adopts a divide-and-combine strategy to leverage the decoupled multi-view information to further improve the generalization performance of the model. The experimental results show that our proposed method significantly outperforms the state-of-the-art algorithms on the benchmark datasets in both multi-task learning and semi-supervised domain adaptation scenarios.

Graph data widely exist in many high-impact applications. Inspired by the success of deep learning in grid-structured data, graph neural network models have been proposed to learn powerful node-level or graph-level representation. However, most of the existing graph neural networks suffer from the following limitations: (1) there is limited analysis regarding the graph convolution properties, such as seed-oriented, degree-aware and order-free; (2) the node's degreespecific graph structure is not explicitly expressed in graph convolution for distinguishing structure-aware node neighborhoods; (3) the theoretical explanation regarding the graph-level pooling schemes is unclear.

In this paper, we propose Task-Adversarial co-Generative Nets (TAGN) for learning from multiple tasks. It aims to address the two fundamental issues of multi-task learning, i.e., domain shift and limited labeled data, in a principled way. To this end, TAGN first learns the task-invariant representations of features to bridge the domain shift among tasks. Based on the task-invariant features, TAGN generates the plausible examples for each task to tackle the data scarcity issue. In TAGN, we leverage multiple game players to gradually improve the quality of the co-generation of features and examples by using an adversarial strategy. It simultaneously learns the marginal distribution of task-invariant features across different tasks and the joint distributions of examples with labels for each task. The theoretical study shows the desired results: at the equilibrium point of the multi-player game, the feature extractor exactly produces the task-invariant features for different tasks, while both the generator and the classifier perfectly replicate the joint distribution for each task. The experimental results on the benchmark data sets demonstrate the effectiveness of the proposed approach.

In contrast to the massive volume of data, it is often the rare categories that are of great importance in many high impact domains, ranging from financial fraud detection in online transaction networks to emerging trend detection in social networks, from spam image detection in social media to rare disease diagnosis in the medical decision support system. The unique challenges of rare category analysis include: (1) the highly-skewed class-membership distribution; (2) the non-separability nature of the rare categories from the majority classes; (3) the data and task heterogeneity, e.g., the multi-modal representation of examples, and the analysis of similar rare categories across multiple related tasks. This tutorial aims to provide a concise review of state-of-the-art techniques on complex rare category analysis, where the majority classes have a smooth distribution, while the minority classes exhibit a compactness property in the feature space or subspace. In particular, we start with the context, problem definition and unique challenges of complex rare category analysis; then we present a comprehensive overview of recent advances that are designed for this problem setting, from rare category exploration without any label information to the exposition step that characterizes rare examples with a compact representation, from representing rare patterns in a salient embedding space to interpreting the prediction results and providing relevant clues for the end users' interpretation; at last, we will discuss the potential challenges and shed light on the future directions of complex rare category analysis.

The increasing need for labeled data has brought the booming growth of crowdsourcing in a wide range of high-impact real-world applications, such as collaborative knowledge (e.g., data annotations, language translations), collective creativity (e.g., analogy mining, crowdfunding), and reverse Turing test (e.g., CAPTCHA-like systems), etc. In the context of supervised learning, crowdsourcing refers to the annotation procedure where the data items are outsourced and processed by a group of mostly unskilled online workers. Thus, the researchers or the organizations are able to collect large amount of information via the feedback of the crowd in a short time with a low cost.

Many real-world problems exhibit the coexistence of multiple types of heterogeneity, such as view heterogeneity (i.e., multi-view property) and task heterogeneity (i.e., multi-task property). For example, in an image classification problem containing multiple poses of the same object, each pose can be considered as one view, and the detection of each type of object can be treated as one task. Furthermore, in some problems, the data type of multiple views might be different. In a web classification problem, for instance, we might be provided an image and text mixed data set, where the web pages are characterized by both images and texts. A common strategy to solve this kind of problem is to leverage the consistency of views and the relatedness of tasks to build the prediction model. In the context of deep neural network, multitask relatedness is usually realized by grouping tasks at each layer, while multi-view consistency is usually enforced by finding the maximal correlation coefficient between views. However, there is no existing deep learning algorithm that jointly models task and view dual heterogeneity, particularly for a data set with multiple modalities (text and image mixed data set or text and video mixed data set, etc.). In this paper, we bridge this gap by proposing a deep multi-task multi-view learning framework that learns a deep representation for such dual-heterogeneity problems. Empirical studies on multiple real-world data sets demonstrate the effectiveness of our proposed Deep-MTMV algorithm.

The unprecedented amounts of data have catalyzed the trend of combining human insights with machine learning techniques, which facilitate the use of crowdsourcing to enlist label information both effectively and efficiently. One crucial challenge in crowdsourcing is the diverse worker quality, which determines the accuracy of the label information provided by such workers. Motivated by the observations that same set of tasks are typically labeled by the same set of workers, we studied their behaviors across multiple related tasks and proposed an optimization framework for learning from task and worker dual heterogeneity. The proposed method uses a weight tensor to represent the workers’ behaviors across multiple tasks, and seeks to find the optimal solution of the tensor by exploiting its structured information. Then, we propose an iterative algorithm to solve the optimization problem and analyze its computational complexity. To infer the true label of an example, we construct a worker ensemble based on the estimated tensor, whose decisions will be weighted using a set of entropy weight. We also prove that the gradient of the most time-consuming updating block is separable with respect to the workers, which leads to a randomized algorithm with faster speed. Moreover, we extend the learning framework to accommodate to the multi-class setting. Finally, we test the performance of our framework on several datasets, and demonstrate its superiority over state-of-the-art techniques.

With the advancements in many data mining and machine learning tasks, together with the availability of large-scale annotated data sets, there have been an increasing number of off-the-shelf tools for addressing these tasks, like Stanford NLP Toolkit and Caffe Model Zoo. However, many of these tasks are time-evolving in nature due to, e.g., the emergence of new features and the change of class conditional distribution of features. As a result, the off-the-shelf tools are not able to adapt to such changes and will suffer from sub-optimal performance in the target application. In this paper, we propose a generic framework named AOT for adapting the outputs from an off-the-shelf tool to accommodate the changes in the learning task. It considers two major types of changes, i.e., label deficiency and distribution shift, and aims to maximally boost the performance of the off-the-shelf tool in the target domain, with the help of a limited number of target domain labeled examples. Furthermore, we propose an iterative algorithm to solve the resulting optimization problem, and we demonstrate the superior performance of the proposed AOT framework on text and image data sets.

Imbalanced data widely exist in many high-impact applications. An example is in air traffic control, where among all three types of accident causes, historical accident reports with `personnel issues' are much more than the other two types (`aircraft issues' and `environmental issues') combined. Thus, the resulting data set of accident reports is highly imbalanced. On the other hand, this data set can be naturally modeled as a network, with each node representing an accident report, and each edge indicating the similarity of a pair of accident reports. Up until now, most existing work on imbalanced data analysis focused on the classification setting, and very little is devoted to learning the node representations for imbalanced networks. To bridge this gap, in this paper, we first propose Vertex-Diminished Random Walk (VDRW) for imbalanced network analysis. It is significantly different from the existing Vertex Reinforced Random Walk by discouraging the random particle to return to the nodes that have already been visited. This design is particularly suitable for imbalanced networks as the random particle is more likely to visit the nodes from the same class, which is a desired property for learning node representations. Furthermore, based on VDRW, we propose a semi-supervised network representation learning framework named ImVerde for imbalanced networks, where context sampling uses VDRW and the limited label information to create node-context pairs, and balanced-batch sampling adopts a simple under-sampling method to balance these pairs from different classes. Experimental results demonstrate that ImVerde based on VDRW outperforms state-of-the-art algorithms for learning network representations from imbalanced data.

Modeling and characterizing high-order connectivity patterns are essential for understanding many complex systems, ranging from social networks to collaboration networks, from finance to neuroscience. However, existing works on high-order graph clustering assume that the input networks are static. Consequently, they fail to explore the rich high-order connectivity patterns embedded in the network evolutions, which may play fundamental roles in real applications. For example, in financial fraud detection, detecting loops formed by sequenced transactions helps identify money laundering activities; in emerging trend detection, star-shaped structures showing in a short burst may indicate novel research topics in citation networks. In this paper, we bridge this gap by proposing a local graph clustering framework that captures structure-rich subgraphs, taking into consideration the information of high-order structures in temporal networks. In particular, our motif-preserving dynamic local graph cut framework (MOTLOC) is able to model various user-defined temporal network structures and find clusters with minimum conductance in a polylogarithmic time complexity. Extensive empirical evaluations on synthetic and real networks demonstrate the effectiveness and efficiency of our MOTLOC framework.

With the rapid growth of e-tail, the cost to handle returned online orders also increases significantly and has become a major challenge in the e-commerce industry. Accurate prediction of product returns allows e-tailers to prevent problematic transactions in advance. However, the limited existing work for modeling customer online shopping behaviors and predicting their return actions fail to integrate the rich information in the product purchase and return history (e.g., return history, purchase-no-return behavior, and customer/product similarity). Furthermore, the large-scale data sets involved in this problem, typically consisting of millions of customers and tens of thousands of products, also render existing methods inefficient and ineffective at predicting the product returns. To address these problems, in this paper, we propose to use a weighted hybrid graph to represent the rich information in the product purchase and return history, in order to predict product returns. The proposed graph consists of both customer nodes and product nodes, undirected edges reflecting customer return history and customer/product similarity based on their attributes, as well as directed edges discriminating purchase-no-return and no-purchase actions. Based on this representation, we study a random-walk-based local algorithm for predicting product return propensity for each customer, whose computational complexity depends only on the size of the output cluster rather than the entire graph. Such a property makes the proposed local algorithm particularly suitable for processing the large-scale data sets to predict product returns. To test the performance of the proposed techniques, we evaluate the graph model and algorithm on multiple e-commerce data sets, showing improved performance over state-of-the-art methods.

Recent decades have witnessed the rapid growth of E-commerce. In particular, E-tail has provided customers with great convenience by allowing them to purchase retail products anywhere without visiting the actual stores. A recent trend in E-tail is to allow free shipping and hassle-free returns to further attract online customers. However, a downside of such a customer-friendly policy is the rapidly increasing return rate as well as the associated costs of handling returned online orders. Therefore, it has become imperative to take proactive measures for reducing the return rate and the associated cost. Despite the large amount of data available from historical purchase and return records, up until now, the problem of E-tail product return prediction has not attracted much attention from the data mining community.

In the era of big data, it is often the rare categories that are of great interest in many high-impact applications, ranging from financial fraud detection in online transaction networks to emerging trend detection in social networks, from network intrusion detection in computer networks to fault detection in manufacturing. As a result, rare category characterization becomes a fundamental learning task, which aims to accurately characterize the rare categories given limited label information. The unique challenge of rare category characterization, i.e., the non-separability nature of the rare categories from the majority classes, together with the availability of the multi-modal representation of the examples, poses a new research question: how can we learn a salient rare category oriented embedding representation such that the rare examples are well separated from the majority class examples in the embedding space, which facilitates the follow-up rare category characterization?

With the increasing demand for large amount of labeled data, crowdsourcing has been used in many large-scale data mining applications. However, most existing works in crowdsourcing mainly focus on label inference and incentive design. In this paper, we address a different problem of adaptive crowd teaching, which is a sub-area of machine teaching in the context of crowdsourcing. Compared with machines, human beings are extremely good at learning a specific target concept (e.g., classifying the images into given categories) and they can also easily transfer the learned concepts into similar learning tasks. Therefore, a more effective way of utilizing crowdsourcing is by supervising the crowd to label in the form of teaching. In order to perform the teaching and expertise estimation simultaneously, we propose an adaptive teaching framework named JEDI to construct the personalized optimal teaching set for the crowdsourcing workers. In JEDI teaching, the teacher assumes that each learner has an exponentially decayed memory. Furthermore, it ensures comprehensiveness in the learning process by carefully balancing teaching diversity and learner's accurate learning in terms of teaching usefulness. Finally, we validate the effectiveness and efficacy of JEDI teaching in comparison with the state-of-the-art techniques on multiple data sets with both synthetic learners and real crowdsourcing workers.

Networks are natural analytic tools in modeling adversarial activities(e.g., human trafficking, illicit drug production, terrorist financial transaction) using different intelligence data sources. However, such activities are often covert and embedded across multiple domains and contexts. They are generally not detectable and recognizable from the perspective of an isolated network, and only become apparent when multiple networks are analyzed in a joint manner. Thus, one of the main research topics in modeling adversarial activities is to develop effective techniques to align and fuse information from different networks into a unified representation for global analysis. Based on the combined network representation, an equally important research topic is on detecting and matching indicating patterns to recognize the underlining adversarial activities in the integrated network. The focus of this workshop is to gather together the researchers from all relevant fields to share their experience and opinions on graph mining techniques in the era of big data, with emphasis on two fundamental problems - "Connecting the dots" and "finding a needle in a haystack", in the context of graph-based adversarial activity analytics.

Functional data is ubiquitous in many domains, such as healthcare, social media, manufacturing process, sensor networks, and so on. The goal of function-on-function regression is to build a mapping from functional predictors to functional response. In this article, we propose a novel function-on-function regression model based on mode-sparsity regularization. The main idea is to represent the regression coefficient function between predictor and response as the double expansion of basis functions, and then use a mode-sparsity regularization to automatically filter out irrelevant basis functions for both predictors and responses. The proposed approach is further extended to the tensor version to accommodate multiple functional predictors. While allowing the dimensionality of the regression weight matrix or tensor to be relatively large, the mode-sparsity regularized model facilitates the multi-way shrinking of basis functions for each mode. The proposed mode-sparsity regularization covers a wide spectrum of sparse models for function-on-function regression. The resulting optimization problem is challenging due to the non-smooth property of the mode-sparsity regularization. We develop an efficient algorithm to solve the problem, which works in an iterative update fashion, and converges to the global optimum. Furthermore, we analyze the generalization performance of the proposed method and derive an upper bound for the consistency between the recovered function and the underlying true function. The effectiveness of the proposed approach is verified on benchmark functional datasets in various domains.

Rare category identification is an important task in many application domains, ranging from network security, to financial fraud detection, to personalized medicine. These are all applications which require the discovery and characterization of sets of rare but structurally-similar data entities which are obscured within a larger but structurally different dataset. This paper introduces RCLens, a visual analytics system designed to support user-guided rare category exploration and identification. RCLens adopts a novel active learning-based algorithm to iteratively identify more accurate rare categories in response to user-provided feedback. The algorithm is tightly integrated with an interactive visualization-based interface which supports a novel and effective workflow for rare category identification. This paper (1) defines RCLens' underlying active-learning algorithm; (2) describes the visualization and interaction designs, including a discussion of how the designs support user-guided rare category identification; and (3) presents results from an evaluation demonstrating RCLens' ability to support the rare category identification process.

With the emergence of online forums associated with major diseases, such as diabetes mellitus, many patients are increasingly dependent on such disease-specific social networks to gain access to additional resources. Among these patients, it is common for them to stick to one disease-specific social network, although their desired resources might be spread over multiple social networks, such as patients with similar questions and concerns. Motivated by this application, in this paper, we focus on cross network link recommendation, which aims to identify similar users across multiple heterogeneous social networks. The problem setting is different from existing work on cross network link prediction, which either tries to link accounts of the same user from different social networks, or aims to match users with complementary expertise or interest. To approach the problem of cross network link recommendation, we propose to jointly decompose the user-keyword matrices from multiple social networks, while requiring them to share the same topics and user group-topic association matrices. This constraint comes from the fact that social networks dedicated to the same disease tend to share the same topics as well as the interests of users groups in certain topics. Based on this intuition, we construct a generic optimization framework, provide four instantiations and an iterative optimization algorithm with performance analysis. In the experiments, we demonstrate the superiority of the proposed algorithm over state-of-the-art techniques on various real-world data sets.

Many real-world applications are characterized by temporal data collected from multiple modalities, each sampled with a different resolution. Examples include manufacturing processes and financial market prediction. In these applications, an interesting observation is that within the same modality, we often have data from multiple views, thus naturally forming a 2-level hierarchy: with the multiple modalities on the top, and the multiple views at the bottom. For example, in aluminum smelting processes, the multiple modalities include power, noise, alumina feed, etc; and within the same modality such as power, the different views correspond to various voltage, current and resistance control signals and measured responses. For such applications, we aim to address the following challenge, i.e., how can we integrate such multi-modality multi-resolution data to effectively predict the targets of interest, such as bath temperature in aluminum smelting cell and the volatility in financial market. In this paper, for the first time, we simultaneously model the hierarchical data structure and the multi-resolution property via a novel framework named HiMuV. Different from existing work based on multiple views on a single level or a single resolution, the proposed framework is based on the key assumption that the information from different modalities is complementary, whereas the information within the same modality (across different views) is redundant in terms of predicting the targets of interest. Therefore, we introduce an optimization framework where the objective function contains both the prediction loss and a novel regularizer enforcing the consistency among different views within the same modality. To solve this optimization framework, we propose an iterative algorithm based on randomized block coordinate descent. Experimental results on synthetic data, benchmark data, and various real data sets from aluminum smelting processes, and stock market prediction demonstrate the effectiveness and efficiency of the proposed algorithm.

Driven by the dramatic growth of data both in terms of the size and sources, learning from heterogeneous data is emerging as an important research direction for many real applications. One of the biggest challenges of this type of problem is how to meaningfully integrate heterogeneous data to considerably improve the generality and quality of the learning model. In this paper, we first present a unified learning framework that aims to leverage the structural information from two types of data heterogeneity: view heterogeneity (as in multi-view learning) and worker heterogeneity (as in crowdsourcing). The objective follows the principles of view consistency and worker consensus by minimizing the loss term with a regularized prediction tensor. We then propose to relax and solve the optimization framework with an iterative updating method. We also prove that the gradient of the most time-consuming updating block is separable with respect to the workers, which leads to a randomized algorithm with faster speed and better convergence. Finally, we compare the proposed method with several state-of-the-arts and demonstrate its effectiveness on various data sets.

Nowadays, as an intrinsic property of big data, data heterogeneity can be seen in a variety of real-world applications, ranging from security to manufacturing, from healthcare to crowdsourcing. It refers to any inhomogeneity in the data, and can be present in a variety of forms, corresponding to different types of data heterogeneity, such as task/view/instance/oracle heterogeneity. As shown in previous work as well as our own work, learning from data heterogeneity not only helps people gain a better understanding of the large volume of data, but also provides a means to leverage such data for effective predictive modeling. In this paper, along with multiple real applications, we will briefly review state-of-the-art techniques for learning from data heterogeneity, and demonstrate their performance at addressing these real world problems.

The growing importance of functional data has fueled the rapid development of functional data analysis, which treats the infinite-dimensional data as continuous functions rather than discrete, finite-dimensional vectors. On the other hand, heterogeneity is an intrinsic property of functional data due to the variety of sources to collect the data. In this paper, we propose a novel multi-task function-on-function regression approach to model both the functionality and heterogeneity of data. The basic idea is to simultaneously model the relatedness among tasks and correlations among basis functions by using the co-grouping structured sparsity to encourage similar tasks to behave similarly in shrinking the basis functions. The resulting optimization problem is challenging due to the non-smoothness and non-separability of the co-grouping structured sparsity. We present an efficient algorithm to solve the problem, and prove its separability, convexity, and global convergence. The proposed algorithm is applicable to a wide spectrum of structured sparsity regularized techniques, such as structured ℓ2,p norm and structured Schatten p-norm. The effectiveness of the proposed approach is verified on benchmark functional data sets collected from various domains.

User behavior modeling is essential in computational advertisement, which builds users' profiles by tracking their online behaviors and then delivers the relevant ads according to each user's interests and needs. Accurate models will lead to higher targeting accuracy and thus improved advertising performance. Intuitively, similar users tend to have similar behaviors towards the displayed ads (e.g., impression, click, conversion). However, to the best of our knowledge, there is not much previous work that explicitly investigates such similarities of various types of user behaviors, and incorporates them into ad response targeting and prediction, largely due to the prohibitive scale of the problem.

Nowadays, large-scale graph data is being generated in a variety of real-world applications, from social networks to co-authorship networks, from protein-protein interaction networks to road traffic networks. Many existing works on graph mining focus on the vertices and edges, with the first-order Markov chain as the underlying model. They fail to explore the high-order network structures, which are of key importance in many high impact domains. For example, in bank customer personally identifiable information (PII) networks, the star structures often correspond to a set of synthetic identities; in financial transaction networks, the loop structures may indicate the existence of money laundering. In this paper, we focus on mining user-specified high-order network structures and aim to find a structure-rich subgraph which does not break many such structures by separating the subgraph from the rest.

Sentiment analysis has been studied for decades, and it is widely used in many real applications such as media monitoring. In sentiment analysis, when addressing the problem of limited labeled data from the target domain, transfer learning, or domain adaptation, has been successfully applied, which borrows information from a relevant source domain with abundant labeled data to improve the prediction performance in the target domain. The key to transfer learning is how to model the relatedness among different domains. For sentiment analysis, a common practice is to assume similar sentiment polarity for the common keywords shared by different domains. However, existing methods largely overlooked the human factor, i.e., the users who expressed such sentiment. In this paper, we address this problem by explicitly modeling the human factor related to sentiment classification. In particular, we assume that the content generated by the same user across different domains is biased in the same way in terms of the sentiment polarity. In other words, optimistic/pessimistic users demonstrate consistent sentiment patterns, no matter what the context is. To this end, we propose a new graph-based approach named U-Cross, which models the relatedness of different domains via both the shared users and keywords. It is non-parametric and semi-supervised in nature. Furthermore, we also study the problem of shared user selection to prevent ‘negative transfer’. In the experiments, we demonstrate the effectiveness of U-Cross by comparing it with existing state-of-the-art techniques on three real data sets.

Dense subgraphs are fundamental patterns in graphs, and dense subgraph detection is often the key step of numerous graph mining applications. Most of the existing methods aim to find a single subgraph with a high density. However, dense subgraphs at different granularities could reveal more intriguing patterns in the underlying graph. In this paper, we propose to hierarchically detect dense subgraphs. The key idea of our method (HiDDen) is to envision the density of subgraphs as a relative measure to its background (i.e., the subgraph at the coarse granularity). Given that the hierarchical dense subgraph detection problem is essentially a nonconvex quadratic programming problem, we propose effective and efficient alternative projected gradient based algorithms to solve it. The experimental evaluations on real graphs demonstrate that (1) our proposed algorithms find subgraphs with an up to 40% higher density in almost every hierarchy; (2) the densities of different hierarchies exhibit a desirable variety across different granularities; (3) our projected gradient descent based algorithm scales linearly w.r.t the number of edges of the input graph; and (4) our methods are able to reveal interesting patterns in the underlying graphs (e.g., synthetic ID in financial fraud detection).

Nowadays, crowdsourcing has been commonly used to enlist label information both effectively and efficiently. One major challenge in crowdsourcing is the diverse worker quality, which determines the accuracy of the label information provided by such workers. Motivated by the observation that in many crowdsourcing platforms, the same set of workers typically work on the same set of tasks, we propose to model the diverse worker quality by studying their behaviors across multiple related tasks. To this end, we propose an optimization framework named MultiC2 for learning from task and worker dual heterogeneity. It uses a weight tensor to represent the workers' behaviors across multiple tasks, and seeks to find the optimal solution of the tensor by exploiting its structured information. We then propose an iterative algorithm to solve the optimization framework and analyze its computational complexity. To infer the true label of an example, we construct a worker ensemble based on the estimated tensor, whose decisions will be weighted using a set of entropy weight. Finally, we test the performance of MultiC2 on various data sets, and demonstrate its superiority over state-of-the-art crowdsourcing techniques.

Many complex real applications involve the collection of time series data with multiple modalities and of multiple resolutions. For example, in aluminum smelting processes, the recorded process variables typically reflect various aspects of these processes, such as pressure and temperature, and they are often obtained with different time resolutions, such as every 5 minutes and every day. How can we effectively leverage both the multi-modality property and the multi-resolution property of the data for the sake of more accurate prediction of key process indicators (e.g., the cell temperature of the aluminum smelting processes)? Different from existing techniques, which can only model the multi-modality property or the multi-resolution property, in this paper, for the first time, we propose to jointly model the two properties such that the prediction results are consistent across multiple modalities and multiple resolutions. To this end, we construct an optimization framework, which is based on a novel regularizer imposing such consistency. Then, we design an effective and efficient optimization algorithm based on randomized block coordinate descent. Its performance is evaluated on both synthetic and real data sets, outperforming state-of-the-art techniques.

Networks are prevalent in many high impact domains. Moreover, cross-domain interactions are frequently observed in many applications, which naturally form the dependencies between different networks. Such kind of highly coupled network systems are referred to as multi-layered networks, and have been used to characterize various complex systems, including critical infrastructure networks, cyber-physical systems, collaboration platforms, biological systems, and many more. Different from single-layered networks where the functionality of their nodes is mainly affected by within-layer connections, multi-layered networks are more vulnerable to disturbance as the impact can be amplified through cross-layer dependencies, leading to the cascade failure to the entire system. To manipulate the connectivity in multi-layered networks, some recent methods have been proposed based on two-layered networks with specific types of connectivity measures. In this paper, we address the above challenges in multiple dimensions. First, we propose a family of connectivity measures (SUBLINE) that unifies a wide range of classic network connectivity measures. Third, we reveal that the connectivity measures in the SUBLINE family enjoy diminishing returns property, which guarantees a near-optimal solution with linear complexity for the connectivity optimization problem. Finally, we evaluate our proposed algorithm on real data sets to demonstrate its effectiveness and efficiency.

Motivated by real applications, heterogeneous learning has emerged as an important research area, which aims to model the co-existence of multiple types of heterogeneity. In this paper, we propose a HEterogeneous REpresentation learning model with structured Sparsity regularization (HERES) to learn from multiple types of heterogeneity. HERES aims to leverage two kinds of information to build a robust learning system. One is the rich correlations among heterogeneous data such as task relatedness, view consistency, and label correlation. The other is the prior knowledge of the data in the form of, e.g., the soft-clustering of the tasks. HERES is a generic framework for heterogeneous learning, which integrates multi-task, multi-view, and multi-label learning into a principled framework based on representation learning. The objective of HERES is to minimize the reconstruction loss of using the factor matrices to recover the input matrix for heterogeneous data, regularized by the structured sparsity constraint. The resulting optimization problem is challenging due to the non-smoothness and non-separability of structured sparsity. We develop an iterative updating method to solve the problem. Furthermore, we prove that the reformulation of structured sparsity is separable, which leads to a family of efficient and scalable algorithms for solving structured sparsity penalized problems. The experimental results in comparison with state-of-the-art methods demonstrate the effectiveness of the proposed approach.

Functional data is ubiquitous in many domains such as healthcare, social media, manufacturing process, sensor networks, etc. Functional data analysis involves the analysis of data which is treated as infinite-dimensional continuous functions rather than discrete, finite-dimensional vectors. In this paper, we propose a novel function-on-function regression model based on mode-sparsity regularization. The main idea is to represent the regression coefficient function between predictor and response as the double expansion of basis functions, and then use mode-sparsity constraint to automatically filter out the irrelevant basis functions for both predictors and responses. The mode-sparsity regularization covers a wide spectrum of sparse models for function-on-function regression. The resulting optimization problem is challenging due to the non-smooth property of the mode-sparsity. We develop an efficient and convergence-guaranteed algorithm to solve the problem. The effectiveness of the proposed approach is verified on benchmark functional data sets in various domains.

Nowadays, temporal data is generated at an unprecedented speed from a variety of applications, such as wearable devices, sensor networks, wireless networks and etc. In contrast to such large amount of temporal data, it is usually the case that only a small portion of them contains information of interest. For example, for the ECG signals collected by wearable devices, most of them collected from healthy people are normal, and only a small number of them collected from people with certain heart diseases are abnormal. Furthermore, even for the abnormal temporal sequences, the abnormal patterns may only be present in a few time segments and are similar among themselves, forming a rare category of temporal patterns. For example, the ECG signal collected from an individual with a certain heart disease may be normal in most time segments, and abnormal in only a few time segments, exhibiting similar patterns. What is even more challenging is that such rare temporal patterns are often non-separable from the normal ones. Existing works on outlier detection for temporal data focus on detecting either the abnormal sequences as a whole, or the abnormal time segments directly, ignoring the relationship between abnormal sequences and abnormal time segments. Moreover, the abnormal patterns are typically treated as isolated outliers instead of a rare category with self-similarity. In this paper, for the first time, we propose a bi-level (sequence-level/ segment-level) model for rare temporal pattern detection. It is based on an optimization framework that fully exploits the bi-level structure in the data, i.e., the relationship between abnormal sequences and abnormal time segments. Furthermore, it uses sequence-specific simple hidden Markov models to obtain segment-level labels, and leverages the similarity among abnormal time segments to estimate the model parameters. To solve the optimization framework, we propose the unsupervised algorithm BIRAD, and also the semi-supervised version BIRAD-K which learns from a single labeled example. Experimental results on both synthetic and real data sets demonstrate the performance of the proposed algorithms from multiple aspects, outperforming state-of-the-art techniques on both temporal outlier detection and rare category analysis.

Multiple types of heterogeneity including label heterogeneity and feature heterogeneity often co-exist in many real-world data mining applications, such as diabetes treatment classification, gene functionality prediction, and brain image analysis. To effectively leverage such heterogeneity, in this article, we propose a novel graph-based model for Learning with both Label and Feature heterogeneity, namely L 2 F . It models the label correlation by requiring that any two label-specific classifiers behave similarly on the same views if the associated labels are similar, and imposes the view consistency by requiring that view-based classifiers generate similar predictions on the same examples. The objective function for L 2 F is jointly convex. To solve the optimization problem, we propose an iterative algorithm, which is guaranteed to converge to the global optimum. One appealing feature of L 2 F is that it is capable of handling data with missing views and labels. Furthermore, we analyze its generalization performance based on Rademacher complexity, which sheds light on the benefits of jointly modeling the label and feature heterogeneity. Experimental results on various biomedical datasets show the effectiveness of the proposed approach.

Recent years have witnessed data explosion in semiconductor manufacturing due to advances in instrumentation and storage techniques. The large amount of data associated with process variables monitored over time form a rich reservoir of information, which can be used for a variety of purposes, such as anomaly detection, quality control, and fault diagnostics. In particular, following the same recipe for a certain Integrated Circuit device, multiple tools and chambers can be deployed for the production of this device, during which multiple time series can be collected, such as temperature, impedance, gas flow, electric bias, etc. These time series naturally fit into a two-dimensional array (matrix), i.e., each element in this array corresponds to a time series for one process variable from one chamber. To leverage the rich structural information in such temporal data, in this article, we propose a novel framework named C-Struts to simultaneously cluster on the two dimensions of this array. In this framework, we interpret the structural information as a set of constraints on the cluster membership, introduce an auxiliary probability distribution accordingly, and design an iterative algorithm to assign each time series to a certain cluster on each dimension. Furthermore, we establish the equivalence between C-Struts and a generic optimization problem, which is able to accommodate various distance functions. Extensive experiments on synthetic, benchmark, as well as manufacturing datasets demonstrate the effectiveness of the proposed method.

In many real world applications such as image annotation, gene function prediction, and insider threat detection, the data collected from heterogeneous sources often exhibit multiple types of heterogeneity, such as task heterogeneity, view heterogeneity, and label heterogeneity. To address this problem, we propose a Hierarchical Multi-Latent Space (HiMLS) learning framework to jointly model the triple types of heterogeneity. The basic idea is to learn a hierarchical multi-latent space by which we can simultaneously leverage the task relatedness, view consistency and the label correlations to improve the learning performance. We first propose a multi-latent space approach to model the complex heterogeneity, which is then used as a building block to stack up a multi-layer structure in order to learn the hierarchical multi-latent space. In such a way, we can gradually learn the more abstract concepts in the higher level. We present two instantiated models of the generalized framework using different divergence measures. The two-phase learning algorithms are used to train the multi-layer models. We drive the multiplicative update rules for pre-training and fine-tuning in each model, and prove the convergence and correctness of the update methods. The effectiveness of the proposed approach is verified on various data sets.

Given the users from a social network site, who have been tagged with a set of terms, how can we recommend the movies tagged with a completely different set of terms hosted by another website? Given the users from a website dedicated to Type I and Type II diabetes, how can we recommend the discussion threads from another website dedicated to gestational diabetes, where the keywords used in the two websites might be quite diverse? In other words, how can we recommend across heterogeneous domains characterized by barely overlapping feature sets?

Networks appear naturally in many high-impact real-world applications. In an increasingly connected and coupled world, the networks arising from many application domains are often collected from different channels, forming the so-called multi-layered networks, such as cyber-physical systems, organization-level collaboration platforms, critical infrastructure networks and many more. Compared with single-layered networks, multi-layered networks are more vulnerable as even a small disturbance on one supporting layer/network might cause a ripple effect to all the dependent layers, leading to a catastrophic/cascading failure of the entire system. The state-of-the-art has been largely focusing on modeling and manipulating the cascading effect of two-layered interdependent network systems for some specific type of network connectivity measure. This paper generalizes the challenge to multiple dimensions. First, we propose a new data model for multi-layered networks MULAN, which admits an arbitrary number of layers with a much more flexible dependency structure among different layers, beyond the current pair-wise dependency. Second, we unify a wide range of classic network connectivity measures SUBLINE. Third, we show that for any connectivity measure in the SUBLINE family, it enjoys the diminishing returns property which in turn lends itself to a family of provable near-optimal control algorithms with linear complexity. Finally, we conduct extensive empirical evaluations on real network data, to validate the effectiveness of the proposed algorithms.

Data heterogeneity is an intrinsic property of many high impact applications, such as insider threat detection, traffic prediction, brain image analysis, quality control in manufacturing processes, etc. Furthermore, multiple types of heterogeneity (e.g., task/view/instance heterogeneity) often co-exist in these applications, thus pose new challenges to existing techniques, most of which are tailored for a single or dual types of heterogeneity. To address this problem, in this paper, we propose a novel graph-based hybrid approach to simultaneously model multiple types of heterogeneity in a principled framework. The objective is to maximize the smoothness consistency of the neighboring nodes, bag-instance correlation together with task relatedness on the hybrid graphs, and simultaneously minimize the empirical classification loss. Furthermore, we analyze its performance based on Rademacher complexity, which sheds light on the benefits of jointly modeling multiple types of heterogeneity. To solve the resulting non-convex non-smooth problem, we propose an iterative algorithm named M3 Learning, which combines block coordinate descent and the bundle method for optimization. Experimental results on various data sets show the effectiveness of the proposed algorithm.

Rare category detection(RCD) is an important topicin data mining, focusing on identifying the initial examples fromrare classes in imbalanced data sets. This problem becomes more challenging when the data is presented as time-evolving graphs, as used in synthetic ID detection and insider threat detection. Most existing techniques for RCD are designed for static data sets, thus not suitable for time-evolving RCD applications. To address this challenge, in this paper, we first proposetwo incremental RCD algorithms, SIRD and BIRD. They arebuilt upon existing density-based techniques for RCD, andincrementally update the detection models, which provide 'timeflexible' RCD. Furthermore, based on BIRD, we propose amodified version named BIRD-LI to deal with the cases wherethe exact priors of the minority classes are not available. Wealso identify a critical task in RCD named query distribution. Itaims to allocate the limited budget into multiple time steps, suchthat the initial examples from the rare classes are detected asearly as possible with the minimum labeling cost. The proposedincremental RCD algorithms and various query distributionstrategies are evaluated empirically on both synthetic and real data.

In many real world applications such as satellite image analysis, gene function prediction, and insider threat detection, the data collected from heterogeneous sources often exhibit multiple types of heterogeneity, such as task heterogeneity, view heterogeneity, and label heterogeneity. To address this problem, we propose a Hierarchical Multi-Latent Space (HiMLS) learning approach to jointly model the triple types of heterogeneity. The basic idea is to learn a hierarchical multi-latent space by which we can simultaneously leverage the task relatedness, view consistency and the label correlations to improve the learning performance. We first propose a multi-latent space framework to model the complex heterogeneity, which is used as a building block to stack up a multi-layer structure so as to learn the hierarchical multi-latent space. In such a way, we can gradually learn the more abstract concepts in the higher level. Then, a deep learning algorithm is proposed to solve the optimization problem. The experimental results on various data sets show the effectiveness of the proposed approach.

This paper targets the problem of cargo pricing optimization in the air cargo business. Given the features associated with a pair of origination and destination, how can we simultaneously predict both the optimal price for the bid stage and the outcome of the transaction (win rate) in the decision stage? In addition, it is often the case that the matrix representing pairs of originations and destinations has a block structure, i.e., the originations and destinations can be co-clustered such that the predictive models are similar within the same co-cluster, and exhibit significant variation among different co-clusters. How can we uncover the co-clusters of originations and destinations while constructing the dual predictive models for the two stages?

We describe a unified active transfer learning framework called Hierarchical Active Transfer Learning (HATL). HATL exploits cluster structure shared between different data domains to perform transfer learning by imputing labels for unlabeled target data and to generate effective label queries during active learning. The resulting framework is flexible enough to perform not only adaptive transfer learning and accelerated active learning but also unsupervised and semi-supervised transfer learning. We derive an intuitive and useful upper bound on HATL's error when used to infer labels for unlabeled target points. We also present results on synthetic data that confirm both intuition and our analysis. Finally, we demonstrate HATL's empirical effectiveness on a benchmark data set for sentiment classification.

In the era of big data, it is often the case that the self-similar rare categories in a large data set are of great importance, such as the malicious insiders in big organizations, and the IC devices with defects in semiconductor manufacturing. Furthermore, such rare categories often exhibit multiple types of heterogeneity, such as the task heterogeneity, which originates from data collected in multiple domains, and the view heterogeneity, which originates from multiple information sources. Existing methods for learning rare categories mainly focus on the homogeneous settings, i.e., a single task and a single view. In this paper, for the first time, we study complex rare categories with both task and view heterogeneity, and propose a novel optimization framework named M2LID. It introduces a boundary characterization metric to capture the sharp changes in density near the boundary of the rare categories in the feature space, and constructs a graph-based model to leverage both task and view heterogeneity. Furthermore, M2LID integrates them in a way of mutual benefit. We also present an effective algorithm to solve this framework, analyze its performance from various aspects, and demonstrate its effectiveness on both synthetic and real datasets.

Multiple types of heterogeneity, such as label heterogeneity and feature heterogeneity, often co-exist in many real-world data mining applications, such as news article categorization, gene functionality prediction. To effectively leverage such heterogeneity, in this paper, we propose a novel graph-based framework for Learning with both Label and Feature heterogeneities, namely L2F. It models the label correlation by requiring that any two label-specific classifiers behave similarly on the same views if the associated labels are similar, and imposes the view consistency by requiring that view-based classifiers generate similar predictions on the same examples. To solve the resulting optimization problem, we propose an iterative algorithm, which is guaranteed to converge to the global optimum. Furthermore, we analyze its generalization performance based on Rademacher complexity, which sheds light on the benefits of jointly modeling the label and feature heterogeneity. Experimental results on various data sets show the effectiveness of the proposed approach.

Recent years have witnessed data explosion in semiconductor manufacturing due to advances in instrumentation and storage techniques. In particular, following the same recipe for a certain IC device, multiple tools and chambers can be deployed for the production of this device, during which multiple time series can be collected, such as temperature, impedance, gas flow, electric bias, etc. These time series naturally fit into a two-dimensional array (matrix), i.e., Each element in this array corresponds to a time series for one process variable from one chamber. To leverage the rich structural information in such temporal data, in this paper, we propose a novel framework named C-Struts to simultaneously cluster on the two dimensions of this array. In this framework, we interpret the structural information as a set of constraints on the cluster membership, introduce an auxiliary probability distribution accordingly, and design an iterative algorithm to assign each time series to a certain cluster on each dimension. To the best of our knowledge, we are the first to address this problem. Extensive experiments on benchmark and manufacturing data sets demonstrate the effectiveness of the proposed method.

Traditional data mining techniques are designed to model a single type of heterogeneity, such as multi-task learning for modeling task heterogeneity, multi-view learning for modeling view heterogeneity, etc. Recently, a variety of real applications emerged, which exhibit dual heterogeneity, namely both task heterogeneity and view heterogeneity. Examples include insider threat detection across multiple organizations, web image classification in different domains, etc. Existing methods for addressing such problems typically assume that multiple tasks are equally related and multiple views are equally consistent, which limits their application in complex settings with varying task relatedness and view consistency. In this paper, we advance state-of-the-art techniques by adaptively modeling task relatedness and view consistency via a nonparametric Bayes model: we model task relatedness using normal penalty with sparse covariances, and view consistency using matrix Dirichlet process. Based on this model, we propose the NOBLE algorithm using an efficient Gibbs sampler. Experimental results on multiple real data sets demonstrate the effectiveness of the proposed algorithm.

Most existing works on multi-task learning (MTL) assume the same input space for different tasks. In this paper, we address a general setting where different tasks have heterogeneous input spaces. This setting has a lot of potential applications, yet it poses new algorithmic challenges - how can we link seemingly uncorrelated tasks to mutually boost their learning performance? Our key observation is that in many real applications, there might exist some correspondence among the inputs of different tasks, which is referred to as pivots. For such applications, we first propose a learning scheme for multiple tasks and analyze its generalization performance. Then we focus on the problems where only a limited number of the pivots are available, and propose a general framework to leverage the pivot information. The idea is to map the heterogeneous input spaces to a common space, and construct a single prediction model in this space for all the tasks. We further propose an effective optimization algorithm to find both the mappings and the prediction model. Experimental results demonstrate its effectiveness, especially with very limited number of pivots.

Multiple Instance Learning (MIL) methods generally represent each example as a collection of instances such that the features for local objects can be better captured, whereas traditional learning methods typically extract a global feature vector for each example as an integral part. However, there is limited research work on which of the two learning scenarios performs better. This paper proposes a novel framework – \emphMultiple Instance LEArning with Global Embedding (MILEAGE), in which the global feature vectors for traditional learning methods are integrated into the MIL setting. MILEAGE can leverage the benefits derived from both learning settings. Within the proposed framework, a large margin method is formulated. In particular, the proposed method adaptively tunes the weights on the two different kinds of feature representations (i.e., global and multiple instance) for each example and trains the classifier simultaneously. An alternative algorithm is proposed to solve the resulting optimization problem, which extends the bundle method to the non-convex case. Some important properties of the proposed method, such as the convergence rate and the generalization error rate, are analyzed. A series of experiments have been conducted to demonstrate the advantages of the proposed method over several state-of-the-art multiple instance and traditional learning methods.

In Multiple Instance Learning (MIL), each entity is normally expressed as a set of instances. Most of the current MIL methods only deal with the case when each instance is represented by one type of features. However, in many real world applications, entities are often described from several different information sources/views. For example, when applying MIL to image categorization, the characteristics of each image can be derived from both its RGB features and SIFT features. Previous research work has shown that, in traditional learning methods, leveraging the consistencies between different information sources could improve the classification performance drastically.

In many real applications, the input data are naturally expressed as tensors, such as virtual metrology in semiconductor manufacturing, face recognition and gait recognition in computer vision, etc. In this paper, we propose a general optimization framework for dealing with tensor inputs. Most existing methods for supervised tensor learning use only rank-one weight tensors in the linear model and cannot readily incorporate domain knowledge. In our framework, we obtain the weight tensor in a hierarchical way — we first approximate it by a low-rank tensor, and then estimate the low-rank approximation using the prior knowledge from various sources, e.g., different domain experts. This is motivated by wafer quality prediction in semiconductor manufacturing. Furthermore, we propose an effective algorithm named H-MOTE for solving this framework, which is guaranteed to converge. The time complexity of H-MOTE is linear with respect to the number of examples as well as the size of the weight tensor. Experimental results show the superiority of H-MOTE over state-of-the-art techniques on both synthetic and real data sets.

Many real problems of multi-task learning exhibit hierarchical task relatedness. In other words, the tasks are partitioned into multiple groups. Different tasks within the same group are related on the task-level, whereas different groups are related on the group-level. For example, in semiconductor manufacturing, the problem of wafer quality prediction can be considered as hierarchical multi-task learning, where each task corresponds to a single side of a chamber with side-level relatedness, and a group of tasks corresponds to a chamber of multiple sides with chamber-level relatedness. Motivated by this application, in this paper, we propose an optimization framework for hierarchical multi-task learning, which partitions all the input features into 2 sets based on their characteristics, and models task-level and group-level relatedness by imposing different constraints on the coefficient vectors of the 2 sets. This is different from existing work on task clustering where the goal is to uncover the grouping of tasks, the tasks do not exhibit group-level relatedness, and the input features are not discriminated in the prediction model to model task-level and group-level relatedness. To solve this framework, we propose the \hear\ algorithm based on block coordinate descent, and demonstrate its effectiveness on both synthetic and real data sets from domains of semiconductor manufacturing and document classification.

Diversified ranking is a fundamental task in machine learning. It is broadly applicable in many real world problems, e.g., information retrieval, team assembling, product search, etc. In this paper, we consider a generic setting where we aim to diversify the top-k ranking list based on an arbitrary relevance function and an arbitrary similarity function among all the examples. We formulate it as an optimization problem and show that in general it is NP-hard. Then, we show that for a large volume of the parameter space, the proposed objective function enjoys the diminishing returns property, which enables us to design a scalable, greedy algorithm to find the near-optimal solution. Experimental results on real data sets demonstrate the effectiveness of the proposed algorithm.

In this paper, we consider the multi-task sparse learning problem under the assumption that the dimensionality diverges with the sample size. The traditional l1/l2 multi-task lasso does not enjoy the oracle property unless a rather strong condition is enforced. Inspired by adaptive lasso, we propose a multi-stage procedure, adaptive multi-task lasso, to simultaneously conduct model estimation and variable selection across different tasks. Motivated by adaptive elastic-net, we further propose the adaptive multi-task elastic-net by adding another quadratic penalty to address the problem of collinearity. When the number of tasks is fixed, under weak assumptions, we establish the asymptotic oracle property for the proposed adaptive multi-task sparse learning methods including both adaptive multi-task lasso and elastic-net. In addition to the desirable asymptotic property, we show by simulations that adaptive sparse learning methods also achieve much improved finite sample performance. As a case study, we apply adaptive multi-task elastic-net to a cognitive science problem, where one wants to discover a compact semantic basis for predicting fMRI images. We show that adaptive multi-task sparse learning methods achieve superior performance and provide some insights into how the brain represents meanings of words.

Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.

Transfer learning has been proposed to address the problem of scarcity of labeled data in the target domain by leveraging the data from the source domain. In many real world applications, data is often represented from different perspectives, which correspond to multiple views. For example, a web page can be described by its contents and its associated links. However, most existing transfer learning methods fail to capture the multi-view {nature}, and might not be best suited for such applications.

Rare categories abound and their characterization has heretofore received little attention. Fraudulent banking transactions, network intrusions, and rare diseases are examples of rare classes whose detection and characterization are of high value. However, accurate characterization is challenging due to high-skewness and non-separability from majority classes, e.g., fraudulent transactions masquerade as legitimate ones. This paper proposes the RACH algorithm by exploring the compactness property of the rare categories. It is based on an optimization framework which encloses the rare examples by a minimum-radius hyper ball. The framework is then converted into a convex optimization problem, which is in turn effectively solved in its dual form by the projected sub gradient method. RACH can be naturally kernelized. Experimental results validate the effectiveness of RACH.

Rare category analysis is of key importance both in theory and in practice. Previous research work focuses on supervised rare category analysis, such as rare category detection and rare category classification. In this paper, for the first time, we address the challenge of unsupervised rare category analysis, including feature selection and rare category selection. We propose to jointly deal with the two correlated tasks so that they can benefit from each other. To this end, we design an optimization framework which is able to co-select the relevant features and the examples from the rare category (a.k.a. the minority class). It is well justified theoretically. Furthermore, we develop the Partial Augmented Lagrangian Method (PALM) to solve the optimization problem. Experimental results on both synthetic and real data sets show the effectiveness of the proposed method.

Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. In this paper, we propose a graph-based transfer learning framework. It propagates the label information from the source domain to the target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bi-partite graph. Our framework is semi-supervised and non-parametric in nature and thus more flexible. We also develop an iterative algorithm so that our framework is scalable to large-scale applications. It enjoys the theoretical property of convergence. Compared with existing transfer learning methods, the proposed framework propagates the label information to both the features irrelevant to the source domain and the unlabeled examples in the target omain via the common features in a principled way. Experimental results on 3 real data sets demonstrate the effectiveness of our algorithm.

Rare category detection is an open challenge in machine learning. It plays the central role in applications such as detecting new financial fraud patterns, detecting new network malware, and scientific discovery. In such cases rare categories are hidden among huge volumes of normal data and observations. In this paper, we propose a new method for rare category detection named SEDER, which requires no prior information about the data set. It implicitly performs semiparametric density estimation using specially designed exponentially families, and then picks the examples for labeling where the neighborhood density changes the most. SEDER can work in the cases where the data is not separable. Its unique feature over all existing methods lies in its prior-free nature, i.e. it does not require any prior information about the data set (e.g. the number of classes, the proportion of the different classes, etc.). Therefore, it is more suitable for real applications. Experimental results on both synthetic and real data sets demonstrate the superiority of SEDER.

Rare category detection is the task of identifying examples from rare classes in an unlabeled data set. It is an open challenge in machine learning and plays key roles in real applications such as financial fraud detection, network intrusion detection, astronomy, spam image detection, etc. In this paper, we develop a new graph-based method for rare category detection named GRADE. It makes use of the global similarity matrix motivated by the manifold ranking algorithm, which results in more compact clusters for the minority classes; by selecting examples from the regions where probability density changes the most, it relaxes the assumption that the majority classes and the minority classes are separable. Furthermore, when detailed information about the data set is not available, we develop a modified version of GRADE named GRADE-LI, which only needs an upper bound on the proportion of each minority class as input. Besides working with data with structured features, both GRADE and GRADE-LI can also work with graph data, which can not be handled by existing rare category detection methods. Experimental results on both synthetic and real data sets demonstrate the effectiveness of the GRADE and GRADE-LI algorithms.

Rare category detection is an open challenge for active learning, especially in the de-novo case (no labeled examples), but of significant practical importance for data mining - e.g. detecting new financial transaction fraud patterns, where normal legitimate transactions dominate. This paper develops a new method for detecting an instance of each minority class via an unsupervised local-density-differential sampling strategy. Essentially a variable-scale nearest neighbor process is used to optimize the probability of sampling tightly-grouped minority classes, subject to a local smoothness assumption of the majority class. Results on both synthetic and real data sets are very positive, detecting each minority class with only a frac- tion of the actively sampled points required by random sampling and by Pelleg’s Interleave method, the prior best technique in the sparse literature on this topic.

In this paper, we propose a general transductive learning framework named generalized manifold-ranking-based image retrieval (gMRBIR) for image retrieval. Comparing with an existing transductive learning method named MRBIR [12], our method could work well whether or not the query image is in the database; thus, it is more applicable for real applications. Given a query image, gMRBIR first initializes a pseudo seed vector based on neighborhood relationship and then spread its scores via manifold ranking to all the unlabeled images in the database. Furthermore, in gMRBIR, we also make use of relevance feedback and active learning to refine the retrieval result so that it converges to the query concept as fast as possible. Systematic experiments on a general-purpose image database consisting of 5,000 Corel images demonstrate the superiority of gMRBIR over state-of-the-art techniques.

In this paper, an optimization based learning method is proposed for image retrieval from graph model point of view. Firstly, image retrieval is formulated as a regularized optimization problem, which simultaneously considers the constraints from low-level feature, online relevance feedback and offline semantic information. Then, the global optimal solution is developed in both closed form and iterative form, providing that the latter converges to the former. The proposed method is unified in the senses that 1) it makes use of the information from various aspects in a global optimization manner so that the retrieval performance might be maximally improved; 2) it provides a natural way to support two typical query scenarios in image retrieval. The proposed method has a solid mathematical ground. Systematic experimental results on a general-purpose image database demonstrate that it achieves significant improvements over existing methods.

To maximally improve the precision among top-ranked images returned by a Web image search engine without putting extra burden on the user, we propose in this paper a novel co-ranking framework which re-ranks the retrieved images to move the irrelevant ones to the tail of the list. The characteristics of the proposed framework can be summarized as follows: (1) making use of the decisions from multi-view of images to boost retrieval performance; (2) generalizing present multi-view algorithms which need labeled data for initialization to the unsupervised case so that no extra interaction is required. To implement the framework, we use one-class support vector machines to train the basic learner, and propose different schemes for combination. Experimental results demonstrate the effectiveness of the proposed framework.

To better understand the content of multimedia, a lot of research efforts have been made on how to learn from multi-modal feature. In this paper, it is studied from a graph point of view: each kind of feature from one modality is represented as one independent graph; and the learning task is formulated as inferring from the constraints in every graph as well as supervision information (if available). For semi-supervised learning, two different fusion schemes, namely linear form and sequential form, are proposed. For each scheme, it is derived from optimization point of view; and further justified from two sides: similarity propagation and Bayesian interpretation. By doing so, we reveal the regular optimization nature, transductive learning nature as well as prior fusion nature of the proposed schemes, respectively. Moreover, the proposed method can be easily extended to unsupervised learning, including clustering and embedding. Systematic experimental results validate the effectiveness of the proposed method.

In this paper, a novel learning based method is proposed for No-Reference image quality assessment. Instead of examining the exact prior knowledge for the given type of distortion and finding a suitable way to represent it, our method aims to directly get the quality metric by means of learning. At first, some training examples are prepared for both high-quality and low-quality classes; then a binary classifier is built on the training set; finally the quality metric of an un-labeled example is denoted by the extent to which it belongs to these two classes. Different schemes to acquire examples from a given image, to build the binary classifier and to model the quality metric are proposed and investigated. While most existing methods are tailored for some specific distortion type, the proposed method might provide a general solution for No-Reference image quality assessment. Experimental results on JPEG and JPEG2000 compressed images validate the effectiveness of the proposed method.

In this paper, we first apply the theory of wallpaper groups to natural images and extract a novel feature to depict the symmetry property of natural images. The original proposed algorithm takes autocorrelation and correlation as a preprocessing step, which is very time-consuming. Through further analysis, we develop a set of schemes to accelerate this algorithm. Experimental results demonstrate that in performing content-based image retrieval, the proposed symmetry feature outperforms the wavelet feature, which is a widely accepted descriptor of texture, and water-filling feature. The accelerated version of the algorithm improves the processing speed by a large margin while it produces little degradation in retrieval performance.

In this paper, we propose a novel transductive learning framework named manifold-ranking based image retrieval (MRBIR). Given a query image, MRBIR first makes use of a manifold ranking algorithm to explore the relationship among all the data points in the feature space, and then measures relevance between the query and all the images in the database accordingly, which is different from traditional similarity metrics based on pair-wise distance. In relevance feedback, if only positive examples are available, they are added to the query set to improve the retrieval result; if examples of both labels can be obtained, MRBIR discriminately spreads the ranking scores of positive and negative examples, considering the asymmetry between these two types of images. Furthermore, three active learning methods are incorporated into MRBIR, which select images in each round of relevance feedback according to different principles, aiming to maximally improve the ranking result. Experimental results on a general-purpose image database show that MRBIR attains a significant improvement over existing systems from all aspects.