HeGCL: Advance Self-Supervised Learning in Heterogeneous Graph-Level Representation.

Aspect-based sentiment analysis aims to predict sentiment polarities of given aspects in text. Most current approaches employ attention-based neural methods to capture semantic relationships between aspects and words in one sentence. However, these methods ignore the fact that sentences with the same aspect and sentiment polarity often share the structure and semantic information in a domain, which leads to lower model performance. To mitigate this problem, we propose a heterogeneous aspect graph neural network (HAGNN) to learn the structure and semantic knowledge from intersentence relationships. Our model is a heterogeneous graph neural network since it contains three different kinds of nodes: word nodes, aspect nodes, and sentence nodes. These nodes can pass structure and semantic information between each other and update their embeddings to improve the performance of our model. To the best of our knowledge, we are the first to use a heterogeneous graph to capture relationships between sentences and aspects. The experimental results on five public datasets show the effectiveness of our model outperforming some state-of-the-art models.

Heterogeneous graph neural networks (HGNNs) have the powerful capability to embed rich structural and semantic information of a heterogeneous graph into node representations. Existing HGNNs inherit many mechanisms from graph neural networks (GNNs) designed for homogeneous graphs, especially the attention mechanism and the multi-layer structure. These mechanisms bring excessive complexity, but seldom work studies whether they are really effective on heterogeneous graphs. In this paper, we conduct an in-depth and detailed study of these mechanisms and propose the Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN). To easily capture structural information, SeHGNN pre-computes the neighbor aggregation using a light-weight mean aggregator, which reduces complexity by removing overused neighbor attention and avoiding repeated neighbor aggregation in every training epoch. To better utilize semantic information, SeHGNN adopts the single-layer structure with long metapaths to extend the receptive field, as well as a transformer-based semantic fusion module to fuse features from different metapaths. As a result, SeHGNN exhibits the characteristics of a simple network structure, high prediction accuracy, and fast training speed. Extensive experiments on five real-world heterogeneous graphs demonstrate the superiority of SeHGNN over the state-of-the-arts on both accuracy and training speed.

Heterogeneous graphs are an essential structure that models real-world data through different types of nodes and relationships between them, including multimodality, which comprises different types of data such as text, image, and audio. Graph Neural Networks (GNNs) are a prominent graph representation learning method that takes advantage of the graph structure and its attributes that, when applied to the multimodal heterogeneous graph, learn a unique semantic space for the different modalities. Consequently, it allows multimodal fusion through simple operators such as sum, average, or multiplication, generating unified representations considering the supplementary and complementarity relationships between the modalities. In multimodal heterogeneous graphs, the labeling process tends to be even more costly due to the multiple modalities analyzed, in addition to the imbalance of classes inherent to some applications. In order to overcome these problems in applications that comprise a class of interest, One-Class Learning (OCL) is used. Given the lack of studies on multimodal early fusion in heterogeneous graphs for OCL tasks, we proposed a method based on unsupervised GNN for heterogeneous graphs and evaluated different early fusion operators. In this paper, we extend another work by evaluating the behavior of the main GNN convolutions in the method. We highlight that using operators such as average, addition, and subtraction were the best early fusion operators. In addition, GNN layers that do not use an attention mechanism performed better. In this way, we argue for heterogeneous graph neural networks in multimodal using early fusion simple operators instead of well-often-used concatenation and less complex convolutions.

Machine learning (ML) has been widely used for efficient resource allocation (RA) in wireless networks. Although superb performance is achieved on small and simple networks, most existing ML-based approaches are confronted with difficulties when heterogeneity occurs and network size expands. In this paper, specifically focusing on power control/beamforming (PC/BF) in heterogeneous device-to-device (D2D) networks, we propose a novel unsupervised learning-based framework named heterogeneous interference graph neural network (HIGNN) to handle these challenges. First, we characterize diversified link features and interference relations with heterogeneous graphs. Then, HIGNN is proposed to empower each link to obtain its individual transmission scheme after limited information exchange with neighboring links. It is noteworthy that HIGNN is scalable to wireless networks of growing sizes with robust performance after trained on small-sized networks. Numerical results show that compared with state-of-the-art benchmarks, HIGNN achieves much higher execution efficiency while providing strong performance.

Graph neural network, as a powerful graph representation technique based on deep learning, has shown superior performance and attracted considerable research interest. However, it has not been fully considered in graph neural network for heterogeneous graph which contains different types of nodes and links. The heterogeneity and rich semantic information bring great challenges for designing a graph neural network for heterogeneous graph. Recently, one of the most exciting advancements in deep learning is the attention mechanism, whose great potential has been well demonstrated in various areas. In this paper, we first propose a novel heterogeneous graph neural network based on the hierarchical attention, including node-level and semantic-level attentions. Specifically, the node-level attention aims to learn the importance between a node and its metapath based neighbors, while the semantic-level attention is able to learn the importance of different meta-paths. With the learned importance from both node-level and semantic-level attention, the importance of node and meta-path can be fully considered. Then the proposed model can generate node embedding by aggregating features from meta-path based neighbors in a hierarchical manner. Extensive experimental results on three real-world heterogeneous graphs not only show the superior performance of our proposed model over the state-of-the-arts, but also demonstrate its potentially good interpretability for graph analysis.

- Heterogeneous graph learning has emerged as a powerful paradigm for modeling complex relational data involving multiple types of nodes and edges. Unlike homogeneous graphs, heterogeneous graphs capture rich semantic relationships across diverse entities, enabling more expressive representations. This paper presents a comprehensive and structured review of heterogeneous graph learning, covering foundational theories, classical approaches, heterogeneous graph neural networks (HGNNs), and transformer-based architectures. Recent advances from 2020 to 2025, including multimodal, dynamic, and contrastive learning frameworks, are also discussed. Furthermore, we analyze experimental benchmarks, applications across domains, and key challenges such as scalability and interpretability. Finally, future research directions are outlined to guide advancements in this rapidly evolving field. Key Words: heterogeneous graphs, graph neural networks, HGNN, graph transformers, representation learning, knowledge graphs, deep learning

To address the challenges faced by existing artificial intelligence methods in modeling complex heterogeneous biological networks, particularly their limitations in capturing collaborative relationships between nodes and in extracting high-order topological semantic features, we propose a novel drug repositioning prediction method based on dynamic representation learning on heterogeneous graphs. A heterogeneous biological graph that integrates drugs, diseases, and their interaction relationships was constructed, based on which a dynamic gated attention module was designed to extract discriminative topological features of drugs and diseases by incorporating a dynamic graph attention mechanism. A gated residual feature fusion mechanism was developed to precisely integrate structural and semantic information from multiple similarity networks to reduce feature redundancy and information loss, thereby enabling accurate prediction of drug-disease associations. Experiments and case studies conducted on multiple drug datasets related to complex diseases demonstrated that the proposed method outperformed existing mainstream models in drug repositioning prediction. The proposed method can effectively model complex associations in heterogeneous biological networks, enhance the accuracy of drug repositioning prediction, and provide important technical support for precision treatment of complex diseases and development of medical artificial intelligence.

Inductive bias in machine learning (ML) is the set of assumptions describing how a model makes predictions. Different ML-based methods for protein-ligand binding affinity (PLA) prediction have different inductive biases, leading to different levels of generalization capability and interpretability. Intuitively, the inductive bias of an ML-based model for PLA prediction should fit in with biological mechanisms relevant for binding to achieve good predictions with meaningful reasons. To this end, we propose an interaction-based inductive bias to restrict neural networks to functions relevant for binding with two assumptions: 1) A protein-ligand complex can be naturally expressed as a heterogeneous graph with covalent and non-covalent interactions; 2) The predicted PLA is the sum of pairwise atom-atom affinities determined by non-covalent interactions. The interaction-based inductive bias is embodied by an explainable heterogeneous interaction graph neural network (EHIGN) for explicitly modeling pairwise atom-atom interactions to predict PLA from 3D structures. Extensive experiments demonstrate that EHIGN achieves better generalization capability than other state-of-the-art ML-based baselines in PLA prediction and structure-based virtual screening. More importantly, comprehensive analyses of distance-affinity, pose-affinity, and substructure-affinity relations suggest that the interaction-based inductive bias can guide the model to learn atomic interactions that are consistent with physical reality. As a case study to demonstrate practical usefulness, our method is tested for predicting the efficacy of Nirmatrelvir against SARS-CoV-2 variants. EHIGN successfully recognizes the changes in the efficacy of Nirmatrelvir for different SARS-CoV-2 variants with meaningful reasons.

Among various region embedding methods, graph-based region relation learning models stand out, owing to their strong structure representation ability for encoding spatial correlations with graph neural networks. Despite their effectiveness, several key challenges have not been well addressed in existing methods: i) Data noise and missing are ubiquitous in many spatio-temporal scenarios due to a variety of factors. ii) Input spatio-temporal data (e.g., mobility traces) usually exhibits distribution heterogeneity across space and time. In such cases, current methods are vulnerable to the quality of the generated region graphs, which may lead to suboptimal performance. In this paper, we tackle the above challenges by exploring the Automated Spatio-Temporal graph contrastive learning paradigm (AutoST) over the heterogeneous region graph generated from multi-view data sources. Our AutoST framework is built upon a heterogeneous graph neural architecture to capture the multi-view region dependencies with respect to POI semantics, mobility flow patterns and geographical positions. To improve the robustness of our GNN encoder against data noise and distribution issues, we design an automated spatio-temporal augmentation scheme with a parameterized contrastive view generator. AutoST can adapt to the spatio-temporal heterogeneous graph with multi-view semantics well preserved. Extensive experiments for three downstream spatio-temporal mining tasks on several real-world datasets demonstrate the significant performance gain achieved by our AutoST over a variety of baselines. The code is publicly available at https://github.com/HKUDS/AutoST.

Traditional approaches for learning representations from somatic mutation data often rely on the distribution of motif counts, which have limitations, including the use of small sequence contexts and the exclusion of critical information, such as mutation location and interactions among mutations. Recent efforts have leveraged advances in natural language processing (NLP) to improve representation learning from somatic mutations by incorporating larger sequence contexts into shallow networks. More recently, studies have demonstrated the feasibility of leveraging attention mechanisms to learn meaningful representations. However, these methods primarily use supervised or weakly supervised learning paradigms, which are inherently task-specific. The development of general-purpose representations, by harnessing the vast amount of somatic mutation data through self-supervised techniques, remains underexplored. Such representations could drive progress in applications including tumor subtype classification, biomarker discovery, patient stratification, drug response prediction, and pathway enrichment analysis. This study aims to develop a deep, unsupervised learning framework that leverages abundant somatic mutation data to generate robust, general-purpose representations. By adapting recent advances in self-supervised learning, the proposed approach seeks to establish a foundation model capable of capturing meaningful patterns in mutation data for diverse downstream tasks. The framework employs novel tokenization and embedding strategies to encode mutation sequences and utilizes transformer-based architectures to model contextual dependencies. The process of inputting mutation data into a transformer architecture, particularly the tokenization step, is critical for this problem, and we explore various tokenization strategies. Benchmarking these representations will include evaluating their utility across tumor subtype classification, biomarker discovery, patient stratification, drug response prediction, identification of mutational signatures. Although the model is under development, experiments from prior work—limited to mutation identifiers without leveraging sequence motifs, mutation location, or other contextual information—have demonstrated the feasibility of encoding mutational data into meaningful representations using transformer-based architectures and self-supervised learning. Building on these findings, this framework will benchmark its performance on multiple downstream tasks, including tumor subtype classification and patient stratification. Citation Format: Marzieh Haghighi, Gustavo Alonso Arango Argoty, Etai Jacob. Self-supervised representation learning of somatic mutational data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5059.

Predicting the next interaction of a short-term interaction session is a challenging task in session-based recommendation. Almost all existing works rely on item transition patterns, and neglect the impact of user historical sessions while modeling user preference, which often leads to non-personalized recommendation. Additionally, existing personalized session-based recommenders capture user preference only based on the sessions of the current user, but ignore the useful item-transition patterns from other user's historical sessions. To address these issues, we propose a novel Heterogeneous Global Graph Neural Networks (HG-GNN) to exploit the item transitions over all sessions in a subtle manner for better inferring user preference from the current and historical sessions. To effectively exploit the item transitions over all sessions from users, we propose a novel heterogeneous global graph that contains item transitions of sessions, user-item interactions and global co-occurrence items. Moreover, to capture user preference from sessions comprehensively, we propose to learn two levels of user representations from the global graph via two graph augmented preference encoders. Specifically, we design a novel heterogeneous graph neural network (HGNN) on the heterogeneous global graph to learn the long-term user preference and item representations with rich semantics. Based on the HGNN, we propose the Current Preference Encoder and the Historical Preference Encoder to capture the different levels of user preference from the current and historical sessions, respectively. To achieve personalized recommendation, we integrate the representations of the user current preference and historical interests to generate the final user preference representation. Extensive experimental results on three real-world datasets show that our model outperforms other state-of-the-art methods.

Recently, Graph Neural Networks (GNNs) have been widely used for fraud detection. GNNs first generate node embedding by aggregating neighboring information under different relations, and then use the final node embedding to detect the node’s suspiciousness. However, traditional GNNs employing only a single type of aggregator fail to capture neighbor information from multiple perspectives and treating different relations equally inevitably weakens the semantic information of heterogeneous graphs. Meanwhile, expressive ability of GNNs is limited by using conventional concatenating or averaging operations to update the center node. Also, camouflaged entities could damage GNN-based models. To handle these problems, a novel heterogeneous GNN model called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Multiple Aggregators and Feature Interactions Network</i> (MAFI) is proposed in this paper to conduct fraud detection tasks. Concretely, multiple types of aggregators are applied on different relations to aggregate neighbor information and aggregator-level attention is utilized to learn the importance of different aggregators. Also, relation-level attention is leveraged to learn the importance of each relation. Besides, conventional update operations are replaced with vector-wise implicit and explicit feature interactions. Moreover, a trainable neighbor sampler is employed to filter camouflaged fraudsters. Comprehensive experiments on two real-world fraud datasets indicate that the proposed MAFI outperforms existing GNN-based fraud detectors.

Heterogeneous graph-based joint representation learning for users and POIs in location-based social network

Graph mining tasks often suffer from the lack of supervision from labeled information due to the intrinsic sparseness of graphs and the high cost of manual annotation. To alleviate this issue, inspired by recent advances of self-supervised learning (SSL) on computer vision and natural language processing, graph self-supervised learning methods have been proposed and achieved remarkable performance by utilizing unlabeled information. However, most existing graph SSL methods focus on homogeneous graphs, ignoring the ubiquitous heterogeneity of real-world graphs where nodes and edges are of multiple types. Therefore, directly applying existing graph SSL methods to heterogeneous graphs can not fully capture the rich semantics and their correlations in heterogeneous graphs. In light of this, we investigate self-supervised learning on heterogeneous graphs and propose a novel model named Multi-View Self-supervised heterogeneous graph Embedding (MVSE). By encoding information from different views defined by meta-paths and optimizing both intra-view and inter-view contrastive learning tasks, MVSE comprehensively utilizes unlabeled information and learns node embeddings. Extensive experiments are conducted on various tasks to show the effectiveness of the proposed framework.

Deep models trained in supervised mode have achieved remarkable success on a variety of tasks. When labeled samples are limited, self-supervised learning (SSL) is emerging as a new paradigm for making use of large amounts of unlabeled samples. SSL has achieved promising performance on natural language and image learning tasks. Recently, there is a trend to extend such success to graph data using graph neural networks (GNNs). In this survey, we provide a unified review of different ways of training GNNs using SSL. Specifically, we categorize SSL methods into contrastive and predictive models. In either category, we provide a unified framework for methods as well as how these methods differ in each component under the framework. Our unified treatment of SSL methods for GNNs sheds light on the similarities and differences of various methods, setting the stage for developing new methods and algorithms. We also summarize different SSL settings and the corresponding datasets used in each setting. To facilitate methodological development and empirical comparison, we develop a standardized testbed for SSL in GNNs, including implementations of common baseline methods, datasets, and evaluation metrics.