We introduce InfoMLP, an innovative model structured like a Multilayer Perceptron (MLP) for semi-supervised classification of structured data, e.g., graphs. InfoMLP was inspired by our observation that overlapping information between node features and the structure between data points significantly influences the performance gap between feature-only MLPs and advanced graph-based semi-supervised methods, e.g., GNNs. To quantify the overlapping information, we first introduce a tractable metric to quantify the mutual information between node features and graph structure. Based on this, we propose InfoMLP, which seeks to maximize the mutual information between node embeddings derived from the MLP and the structure information. Our info-max objective is split into two sub-objectives: the first is a non-parametric preprocessing step aiming to find the optimal graph-augmented node feature matrix that captures the maximal information about the graph structure, while the second sub-objective is to maximize the mutual information between node embeddings generated from the original node features and those from the graph-augmented node features. Since the message-passing operation is integrated into the preprocessing step, requiring only a single execution per dataset, InfoMLP maintains the same efficiency as a vanilla MLP during both training and testing. We validate the efficacy of our design through experiments on real-world datasets of varying scales supplemented by comprehensive ablation studies. Our results corroborate our analysis and demonstrate the effectiveness of our novel approach.

Mixup has shown considerable success in mitigating the challenges posed by limited labeled data in image classification. By synthesizing samples through the interpolation of features and labels, Mixup effectively addresses the issue of data scarcity. However, it has rarely been explored in graph learning tasks due to the irregularity and connectivity of graph data. Specifically, in node classification tasks, Mixup presents a challenge in creating connections for synthetic data. In this paper, we propose Geometric Mixup (GeoMix), a simple and interpretable Mixup approach leveraging in-place graph editing. It effectively utilizes geometry information to interpolate features and labels with those from the nearby neighborhood, generating synthetic nodes and establishing connections for them. We conduct theoretical analysis to elucidate the rationale behind employing geometry information for node Mixup, emphasizing the significance of locality enhancement-a critical aspect of our method's design. Extensive experiments demonstrate that our lightweight Geometric Mixup achieves state-of-the-art results on a wide variety of standard datasets with limited labeled data. Furthermore, it significantly improves the generalization capability of underlying GNNs across various challenging out-of-distribution generalization tasks. Our code is available at https://github.com/WtaoZhao/geomix.

Out-of-distribution (OOD) generalization has gained increasing attentions for learning on graphs, as graph neural networks (GNNs) often exhibit performance degradation with distribution shifts. The challenge is that distribution shifts on graphs involve intricate interconnections between nodes, and the environment labels are often absent in data. In this paper, we adopt a bottom-up data-generative perspective and reveal a key observation through causal analysis: the crux of GNNs' failure in OOD generalization lies in the latent confounding bias from the environment. The latter misguides the model to leverage environment-sensitive correlations between ego-graph features and target nodes' labels, resulting in undesirable generalization on new unseen nodes. Built upon this analysis, we introduce a conceptually simple yet principled approach for training robust GNNs under node-level distribution shifts, without prior knowledge of environment labels. Our method resorts to a new learning objective derived from causal inference that coordinates an environment estimator and a mixture-of-expert GNN predictor. The new approach can counteract the confounding bias in training data and facilitate learning generalizable predictive relations. Extensive experiment demonstrates that our model can effectively enhance generalization with various types of distribution shifts and yield up to 27.4% accuracy improvement over state-of-the-arts on graph OOD generalization benchmarks.

Cross-Domain Sequential Recommendation (CDSR) methods aim to tackle the data sparsity and cold-start problems present in Single-Domain Sequential Recommendation (SDSR). Existing CDSR works design their elaborate structures relying on overlapping users to propagate the cross-domain information. However, current CDSR methods make closed-world assumptions, assuming fully overlapping users across multiple domains and that the data distribution remains unchanged from the training environment to the test environment. As a result, these methods typically result in lower performance on online real-world platforms due to the data distribution shifts. To address these challenges under open-world assumptions, we design an Adaptive Multi-Interest Debiasing framework for cross-domain sequential recommendation (AMID), which consists of a multi-interest information module (MIM) and a doubly robust estimator (DRE). Our framework is adaptive for open-world environments and can improve the model of most off-the-shelf single-domain sequential backbone models for CDSR. Our MIM establishes interest groups that consider both overlapping and non-overlapping users, allowing us to effectively explore user intent and explicit interest. To alleviate biases across multiple domains, we developed the DRE for the CDSR methods. We also provide a theoretical analysis that demonstrates the superiority of our proposed estimator in terms of bias and tail bound, compared to the IPS estimator used in previous work. To promote related research in the community under open-world assumptions, we collected an industry financial CDSR dataset from Alipay, called "MYbank-CDR". Extensive offline experiments on four industry CDSR scenarios including the Amazon and MYbank-CDR datasets demonstrate the remarkable performance of our proposed approach. Additionally, we conducted a standard A/B test on Alipay, a large-scale financial platform with over one billion users, to validate the effectiveness of our model under open-world assumptions. Code and dataset are available at https://github.com/WujiangXu/AMID.

Graph structure learning is a well-established problem that aims at optimizing graph structures adaptive to specific graph datasets to help message passing neural networks (i.e., GNNs) to yield effective and robust node embeddings. However, the common limitation of existing models lies in the underlying closed-world assumption: the testing graph is the same as the training graph. This premise requires independently training the structure learning model from scratch for each graph dataset, which leads to prohibitive computation costs and potential risks for serious over-fitting. To mitigate these issues, this paper explores a new direction that moves forward to learn a universal structure learning model that can generalize across graph datasets in an open world. We first introduce the mathematical definition of this novel problem setting, and describe the model formulation from a probabilistic data-generative aspect. Then we devise a general framework that coordinates a single graph-shared structure learner and multiple graph-specific GNNs to capture the generalizable patterns of optimal message-passing topology across datasets. The well-trained structure learner can directly produce adaptive structures for unseen target graphs without any fine-tuning. Across diverse datasets and various challenging cross-graph generalization protocols, our experiments show that even without training on target graphs, the proposed model i) significantly outperforms expressive GNNs trained on input (non-optimized) topology, and ii) surprisingly performs on par with state-of-the-art models that independently optimizes adaptive structures for specific target graphs, with notably orders-of-magnitude acceleration for training on the target graph.

Combinatorial drug recommendation involves recommending a personalized combination of medication (drugs) to a patient over his/her longitudinal history, which essentially aims at solving a combinatorial optimization problem that pursues high accuracy under the safety constraint. Among existing learning-based approaches, the association between drug substructures (i.e., a sub-graph of the molecule that contributes to certain chemical effect) and the target disease is largely overlooked, though the function of drugs in fact exhibits strong relevance with particular substructures. To address this issue, we propose a molecular substructure-aware encoding method entitled MoleRec that entails a hierarchical architecture aimed at modeling inter-substructure interactions and individual substructures’ impact on patient’s health condition, in order to identify those substructures that really contribute to healing patients. Specifically, MoleRec learns to attentively pooling over substructure representations which will be element-wisely re-scaled by the model’s inferred relevancy with a patient’s health condition to obtain a prior-knowledge-informed drug representation. We further design a weight annealing strategy for drug-drug-interaction (DDI) objective to adaptively control the balance between accuracy and safety criteria throughout training. Experiments on the MIMIC-III dataset demonstrate that our approach achieves new state-of-the-art performance w.r.t. four accuracy and safety metrics. Our source code is publicly available at https://github.com/yangnianzu0515/MoleRec.

Graph Neural Networks (GNNs) are one of the prominent methods to solve semi-supervised learning on graphs. However, most of the existing GNN models often need sufficient observed data to allow for effective learning and generalization. In real-world scenarios where complete input graph structure and sufficient node labels might not be achieved easily, GNN models would encounter with severe performance degradation. To address this problem, we propose WSGNN, short for weakly-supervised graph neural network. WSGNN is a flexible probabilistic generative framework which harnesses variational inference approach to solve graph semi-supervised learning in a label-structure joint estimation manner. It collaboratively learns task-related new graph structure and node representations through a two-branch network, and targets a composite variational objective derived from underlying data generation distribution concerning the inter-dependence between scarce observed data and massive missing data. Especially, under weakly-supervised low-data regime where labeled nodes and observed edges are both very limited, extensive experimental results on node classification and link prediction over common benchmarks demonstrate the state-of-the-art performance of WSGNN over strong competitors. Concretely, when only 1 label per class and 1% edges are observed on Cora, WSGNN maintains a decent 52.00% classification accuracy, exceeding GCN by 75.6%.

The adversarial attack reveals the vulnerability of deep models by incurring test domain shift, while delusive attack relieves the privacy concern about personal data by injecting malicious noise into the training domain to make data unexploitable. However, beyond their successful applications, the two attacks can be easily defended by adversarial training (AT). While AT is not the panacea, it suffers from poor generalization for robustness. For the limitations of attack and defense, we argue that to fit data well, DNNs can learn the spurious relations between inputs and outputs, which are consequently utilized by the attack and defense and degrade their effectiveness, and DNNs can not easily capture the causal relations like humans to make robust decisions under attacks. In this paper, to better understand and improve attack and defense, we first take a bottom-up perspective to describe the correlations between latent factors and observed data, then analyze the effect of domain shift on DNNs induced by attack and finally develop our causal graph, namely Domain-attack Invariant Causal Model (DICM). Based on DICM, we propose a coherent causal invariant principle, which guides our algorithm design to infer the human-like causal relations. We call our algorithm Domain-attack Invariant Causal Learning (DICE) and the experimental results on two attacks and one defense task verify its effectiveness.

User behavior sequences contain rich information about user interests and are exploited to predict user's future clicking in sequential recommendation. Existing approaches, especially recently proposed deep learning models, often embed a sequence of clicked items into a single vector, i.e., a point in vector space, which suffer from limited expressiveness for complex distributions of user interests with multi-modality and heterogeneous concentration. In this paper, we propose a new representation model, named as Seq2Bubbles, for sequential user behaviors via embedding an input sequence into a set of bubbles each of which is represented by a center vector and a radius vector in embedding space. The bubble embedding can effectively identify and accommodate multi-modal user interests and diverse concentration levels. Furthermore, we design an efficient scheme to compute distance between a target item and the bubble embedding of a user sequence to achieve next-item recommendation. We also develop a self-supervised contrastive loss based on our bubble embeddings as an effective regularization approach. Extensive experiments on four benchmark datasets demonstrate that our bubble embedding can consistently outperform state-of-the-art sequential recommendation models.

Sequential recommendation and information dissemination are two traditional problems for sequential information retrieval. The common goal of the two problems is to predict future user-item interactions based on past observed interactions. The difference is that the former deals with users' histories of clicked items, while the latter focuses on items' histories of infected users.In this paper, we take a fresh view and propose dual sequential prediction models that unify these two thinking paradigms. One user-centered model takes a user's historical sequence of interactions as input, captures the user's dynamic states, and approximates the conditional probability of the next interaction for a given item based on the user's past clicking logs. By contrast, one item-centered model leverages an item's history, captures the item's dynamic states, and approximates the conditional probability of the next interaction for a given user based on the item's past infection records. To take advantage of the dual information, we design a new training mechanism which lets the two models play a game with each other and use the predicted score from the opponent to design a feedback signal to guide the training. We show that the dual models can better distinguish false negative samples and true negative samples compared with single sequential recommendation or information dissemination models. Experiments on four real-world datasets demonstrate the superiority of proposed model over some strong baselines as well as the effectiveness of dual training mechanism between two models.

Social recommendation leverages social information to solve data sparsity and cold-start problems in traditional collaborative filtering methods. However, most existing models assume that social effects from friend users are static and under the forms of constant weights or fixed constraints. To relax this strong assumption, in this paper, we propose dual graph attention networks to collaboratively learn representations for two-fold social effects, where one is modeled by a user-specific attention weight and the other is modeled by a dynamic and context-aware attention weight. We also extend the social effects in user domain to item domain, so that information from related items can be leveraged to further alleviate the data sparsity problem. Furthermore, considering that different social effects in two domains could interact with each other and jointly influence users' preferences for items, we propose a new policy-based fusion strategy based on contextual multi-armed bandit to weigh interactions of various social effects. Experiments on one benchmark dataset and a commercial dataset verify the efficacy of the key components in our model. The results show that our model achieves great improvement for recommendation accuracy compared with other state-of-the-art social recommendation methods.

This paper targets a general popularity prediction problem for event sequence, which has recently gained great attention due to its extensive applications in various domains. Feature driven method and point process method are two basic thinking paradigms to tackle the prediction problem, but both of them suffer from limitations. In this paper, we propose PreNets unifying the two thinking paradigms in an adversarial manner. On one side, feature driven model acts like a 'critic' who aims to discriminate the predicted popularity from the real one based on a set of temporal features from the sequence. On the other side, point process model acts like an 'interpreter' who recognizes the dynamic patterns in sequence to generate a predicted popularity that can fool the 'critic'. Through a Wasserstein learning based two-player game, the training loss of the 'critic' guides the 'interpreter' to better exploit the sequence patterns and enhance prediction, while the 'interpreter' pushes the 'critic' to select effective early features that helps discrimination. This mechanism enables the framework to absorb the advantages of both feature driven and point process methods. Empirical results show that PreNets achieves significant MAPE improvement for both Twitter cascade and Amazon review prediction.

The boom of information technology enables social platforms (like Twitter) to disseminate social content (like news) in an unprecedented rate, which makes early-stage prediction for social content popularity of great practical significance. However, most existing studies assume a long-term observation before prediction and suffer from limited precision for early-stage prediction due to insufficient observation. In this paper, we take a fresh perspective, and propose a novel early pattern aware Bayesian model. The early pattern representation, which stands for early time series normalized on future popularity, can address what we call early-stage indistinctiveness challenge. Then we use an expressive evolving function to fit the time series and estimate three interpretable coefficients characterizing temporal effect of observed series on future evolution. Furthermore, Bayesian network is leveraged to model the probabilistic relations among features, early indicators and early patterns. Experiments on three real-world social platforms (Twitter, Weibo and WeChat) show that under different evaluation metrics, our model outperforms other methods in early-stage prediction and possesses low sensitivity to observation time.