CASPER: Causality-Aware Spatiotemporal Graph Neural Networks for Spatiotemporal Time Series Imputation

Spatiotemporal time series is the foundation of understanding human activities and their impacts, which is usually collected via monitoring sensors placed at different locations. The collected data usually contains missing values due to various failures, which have significant impact on data analysis. To impute the missing values, a lot of methods have been introduced. When recovering a specific data point, most existing methods tend to take into consideration all the information relevant to that point regardless of whether they have a cause-and-effect relationship. During data collection, it is inevitable that some unknown confounders are included, e.g., background noise in time series and non-causal shortcut edges in the constructed sensor network. These confounders could open backdoor paths between the input and output, in other words, they establish non-causal correlations between the input and output. Over-exploiting these non-causal correlations could result in overfitting and make the model vulnerable to noises. In this paper, we first revisit spatiotemporal time series imputation from a causal perspective, which shows the causal relationships among the input, output, embeddings and confounders. Next, we show how to block the confounders via the frontdoor adjustment. Based on the results of the frontdoor adjustment, we introduce a novel Causality-Aware SPatiotEmpoRal graph neural network (CASPER), which contains a novel Spatiotemporal Causal Attention (SCA) and a Prompt Based Decoder (PBD). PBD could reduce the impact of confounders and SCA could discover the sparse causal relationships among embeddings. Theoretical analysis reveals that SCA discovers causal relationships based on the values of gradients. We evaluate Casper on three real-world datasets, and the experimental results show that Casper outperforms the baselines and effectively discovers causal relationships.

Mitigating Feature Gap for Adversarial Robustness by Feature Disentanglement

Deep neural networks are vulnerable to adversarial samples. Adversarial fine-tuning methods aim to enhance adversarial robustness through fine-tuning the naturally pre-trained model in an adversarial training manner. However, we identify that some latent features of adversarial samples are confused by adversarial perturbation and lead to an unexpectedly increasing gap between features in the last hidden layer of natural and adversarial samples. To address this issue, we propose a disentanglement-based approach to explicitly model and further remove the latent features that cause the feature gap. Specifically, we introduce a feature disentangler to separate out the latent features from the features of the adversarial samples, thereby boosting robustness by eliminating the latent features. Besides, we align features in the pre-trained model with features of adversarial samples in the fine-tuned model, to further benefit from the features from natural samples without confusion. Empirical evaluations on three benchmark datasets demonstrate that our approach surpasses existing adversarial fine-tuning methods and adversarial training baselines.

Robust Representation Learning via Asymmetric Negative Contrast and Reverse Attention

Deep neural networks are vulnerable to adversarial noise. Adversarial training (AT) has been demonstrated to be the most effective defense strategy to protect neural networks from being fooled. However, we find AT omits to learning robust features, resulting in poor performance of adversarial robustness. To address this issue, we highlight two characteristics of robust representation: (1) $\bf{exclusion}$: the feature of natural examples keeps away from that of other classes; (2) $\bf{alignment}$: the feature of natural and corresponding adversarial examples is close to each other. These motivate us to propose a generic framework of AT to gain robust representation, by the asymmetric negative contrast and reverse attention. Specifically, we design an asymmetric negative contrast based on predicted probabilities, to push away examples of different classes in the feature space. Moreover, we propose to weight feature by parameters of the linear classifier as the reverse attention, to obtain class-aware feature and pull close the feature of the same class. Empirical evaluations on three benchmark datasets show our methods greatly advance the robustness of AT and achieve state-of-the-art performance. Code is available at <https://github.com/changzhang777/ANCRA>.

Gradient constrained sharpness-aware prompt learning for vision-language models

This paper targets a novel trade-off problem in generalizable prompt learning for vision-language models (VLM), i.e., improving the performance on unseen classes while maintaining the performance on seen classes. Comparing with existing generalizable methods that neglect the seen classes degradation, the setting of this problem is more strict and fits more closely with practical applications. To solve this problem, we start from the optimization perspective, and leverage the relationship between loss landscape geometry and model generalization ability. By analyzing the loss landscapes of the state-of-the-art method and vanilla Sharpness-aware Minimization (SAM) based method, we conclude that the trade-off performance correlates to both loss value and loss sharpness, while each of them is indispensable. However, we find the optimizing gradient of existing methods cannot maintain high relevance to both loss value and loss sharpness during optimization, which severely affects their trade-off performance. To this end, we propose a novel SAM-based method for prompt learning, denoted as Gradient Constrained Sharpness-aware Context Optimization (GCSCoOp), to dynamically constrain the optimizing gradient, thus achieving above two-fold optimization objective simultaneously. Extensive experiments verify the effectiveness of GCSCoOp in the trade-off problem.

Towards Reliable Rare Category Analysis on Graphs via Individual Calibration

Rare categories abound in a number of real-world networks and play a pivotal role in a variety of high-stakes applications, including financial fraud detection, network intrusion detection, and rare disease diagnosis. Rare category analysis (RCA) refers to the task of detecting, characterizing, and comprehending the behaviors of minority classes in a highly-imbalanced data distribution. While the vast majority of existing work on RCA has focused on improving the prediction performance, a few fundamental research questions heretofore have received little attention and are less explored: How confident or uncertain is a prediction model in rare category analysis? How can we quantify the uncertainty in the learning process and enable reliable rare category analysis? To answer these questions, we start by investigating miscalibration in existing RCA methods. Empirical results reveal that state-of-the-art RCA methods are mainly over-confident in predicting minority classes and under-confident in predicting majority classes. Motivated by the observation, we propose a novel individual calibration framework, named CALIRARE, for alleviating the unique challenges of RCA, thus enabling reliable rare category analysis. In particular, to quantify the uncertainties in RCA, we develop a node-level uncertainty quantification algorithm to model the overlapping support regions with high uncertainty; to handle the rarity of minority classes in miscalibration calculation, we generalize the distribution-based calibration metric to the instance level and propose the first individual calibration measurement on graphs named Expected Individual Calibration Error (EICE). We perform extensive experimental evaluations on real-world datasets, including rare category characterization and model calibration tasks, which demonstrate the significance of our proposed framework.

HeroLT: Benchmarking Heterogeneous Long-Tailed Learning

Long-tailed data distributions are prevalent in a variety of domains, including finance, e-commerce, biomedical science, and cyber security. In such scenarios, the performance of machine learning models is often dominated by the head categories, while the learning of tail categories is significantly inadequate. Given abundant studies conducted to alleviate the issue, this work aims to provide a systematic view of long-tailed learning with regard to three pivotal angles: (A1) the characterization of data long-tailedness, (A2) the data complexity of various domains, and (A3) the heterogeneity of emerging tasks. To achieve this, we develop the most comprehensive (to the best of our knowledge) long-tailed learning benchmark named HeroLT, which integrates 13 state-of-the-art algorithms and 6 evaluation metrics on 14 real-world benchmark datasets across 4 tasks from 3 domains. HeroLT with novel angles and extensive experiments (264 in total) enables researchers and practitioners to effectively and fairly evaluate newly proposed methods compared with existing baselines on varying types of datasets. Finally, we conclude by highlighting the significant applications of long-tailed learning and identifying several promising future directions. For accessibility and reproducibility, we open-source our benchmark HeroLT and corresponding results at https://github.com/SSSKJ/HeroLT.

GPatcher: A Simple and Adaptive MLP Model for Alleviating Graph Heterophily

While graph heterophily has been extensively studied in recent years, a fundamental research question largely remains nascent: How and to what extent will graph heterophily affect the prediction performance of graph neural networks (GNNs)? In this paper, we aim to demystify the impact of graph heterophily on GNN spectral filters. Our theoretical results show that it is essential to design adaptive polynomial filters that adapts different degrees of graph heterophily to guarantee the generalization performance of GNNs. Inspired by our theoretical findings, we propose a simple yet powerful GNN named GPatcher by leveraging the MLP-Mixer architectures. Our approach comprises two main components: (1) an adaptive patch extractor function that automatically transforms each node's non-Euclidean graph representations to Euclidean patch representations given different degrees of heterophily, and (2) an efficient patch mixer function that learns salient node representation from both the local context information and the global positional information. Through extensive experiments, the GPatcher model demonstrates outstanding performance on node classification compared with popular homophily GNNs and state-of-the-art heterophily GNNs.

Characterizing Long-Tail Categories on Graphs

Long-tail data distributions are prevalent in many real-world networks, including financial transaction networks, e-commerce networks, and collaboration networks. Despite the success of recent developments, the existing works mainly focus on debiasing the machine learning models via graph augmentation or objective reweighting. However, there is limited literature that provides a theoretical tool to characterize the behaviors of long-tail categories on graphs and understand the generalization performance in real scenarios. To bridge this gap, we propose the first generalization bound for long-tail classification on graphs by formulating the problem in the fashion of multi-task learning, i.e., each task corresponds to the prediction of one particular category. Our theoretical results show that the generalization performance of long-tail classification is dominated by the range of losses across all tasks and the total number of tasks. Building upon the theoretical findings, we propose a novel generic framework Tail2Learn to improve the performance of long-tail categories on graphs. In particular, we start with a hierarchical task grouping module that allows label-limited classes to benefit from the relevant information shared by other classes; then, we further design a balanced contrastive learning module to balance the gradient contributions of head and tail classes. Finally, extensive experiments on various real-world datasets demonstrate the effectiveness of Tail2Learn in capturing long-tail categories on graphs.

Dynamic Transfer Learning across Graphs

Transferring knowledge across graphs plays a pivotal role in many high-stake domains, ranging from transportation networks to e-commerce networks, from neuroscience to finance. To date, the vast majority of existing works assume both source and target domains are sampled from a universal and stationary distribution. However, many real-world systems are intrinsically dynamic, where the underlying domains are evolving over time. To bridge the gap, we propose to shift the problem to the dynamic setting and ask: given the label-rich source graphs and the label-scarce target graphs observed in previous T timestamps, how can we effectively characterize the evolving domain discrepancy and optimize the generalization performance of the target domain at the incoming T+1 timestamp? To answer the question, for the first time, we propose a generalization bound under the setting of dynamic transfer learning across graphs, which implies the generalization performance is dominated by domain evolution and domain discrepancy between source and target domains. Inspired by the theoretical results, we propose a novel generic framework DyTrans to improve knowledge transferability across dynamic graphs. In particular, we start with a transformer-based temporal encoding module to model temporal information of the evolving domains; then, we further design a dynamic domain unification module to efficiently learn domain-invariant representations across the source and target domains. Finally, extensive experiments on various real-world datasets demonstrate the effectiveness of DyTrans in transferring knowledge from dynamic source domains to dynamic target domains.