Dynamic graph neural networks (DGNNs) are increasingly pervasive in exploiting spatio-temporal patterns on dynamic graphs. However, existing works fail to generalize under distribution shifts, which are common in real-world scenarios. As the generation of dynamic graphs is heavily influenced by latent environments, investigating their impacts on the out-of-distribution (OOD) generalization is critical. However, it remains unexplored with the following two major challenges: (1) How to properly model and infer the complex environments on dynamic graphs with distribution shifts? (2) How to discover invariant patterns given inferred spatio-temporal environments? To solve these challenges, we propose a novel Environment-Aware dynamic Graph LEarning (EAGLE) framework for OOD generalization by modeling complex coupled environments and exploiting spatio-temporal invariant patterns. Specifically, we first design the environment-aware EA-DGNN to model environments by multi-channel environments disentangling. Then, we propose an environment instantiation mechanism for environment diversification with inferred distributions. Finally, we discriminate spatio-temporal invariant patterns for out-of-distribution prediction by the invariant pattern recognition mechanism and perform fine-grained causal interventions node-wisely with a mixture of instantiated environment samples. Experiments on real-world and synthetic dynamic graph datasets demonstrate the superiority of our method against state-of-the-art baselines under distribution shifts. To the best of our knowledge, we are the first to study OOD generalization on dynamic graphs from the environment learning perspective.
Recently, continual graph learning has been increasingly adopted for diverse graph-structured data processing tasks in non-stationary environments. Despite its promising learning capability, current studies on continual graph learning mainly focus on mitigating the catastrophic forgetting problem while ignoring continuous performance improvement. To bridge this gap, this article aims to provide a comprehensive survey of recent efforts on continual graph learning. Specifically, we introduce a new taxonomy of continual graph learning from the perspective of overcoming catastrophic forgetting. Moreover, we systematically analyze the challenges of applying these continual graph learning methods in improving performance continuously and then discuss the possible solutions. Finally, we present open issues and future directions pertaining to the development of continual graph learning and discuss how they impact continuous performance improvement.
Graph kernels used to be the dominant approach to feature engineering for structured data, which are superseded by modern GNNs as the former lacks learnability. Recently, a suite of Kernel Convolution Networks (KCNs) successfully revitalized graph kernels by introducing learnability, which convolves input with learnable hidden graphs using a certain graph kernel. The random walk kernel (RWK) has been used as the default kernel in many KCNs, gaining increasing attention. In this paper, we first revisit the RWK and its current usage in KCNs, revealing several shortcomings of the existing designs, and propose an improved graph kernel RWK+, by introducing color-matching random walks and deriving its efficient computation. We then propose RWK+CN, a KCN that uses RWK+ as the core kernel to learn descriptive graph features with an unsupervised objective, which can not be achieved by GNNs. Further, by unrolling RWK+, we discover its connection with a regular GCN layer, and propose a novel GNN layer RWK+Conv. In the first part of experiments, we demonstrate the descriptive learning ability of RWK+CN with the improved random walk kernel RWK+ on unsupervised pattern mining tasks; in the second part, we show the effectiveness of RWK+ for a variety of KCN architectures and supervised graph learning tasks, and demonstrate the expressiveness of RWK+Conv layer, especially on the graph-level tasks. RWK+ and RWK+Conv adapt to various real-world applications, including web applications such as bot detection in a web-scale Twitter social network, and community classification in Reddit social interaction networks.
Scalable graph neural networks (GNNs) have emerged as a promising technique, which exhibits superior predictive performance and high running efficiency across numerous large-scale graph-based web applications. However, (i) Most scalable GNNs tend to treat all nodes in graphs with the same propagation rules, neglecting their topological uniqueness; (ii) Existing node-wise propagation optimization strategies are insufficient on web-scale graphs with intricate topology, where a full portrayal of nodes' local properties is required. Intuitively, different nodes in web-scale graphs possess distinct topological roles, and therefore propagating them indiscriminately or neglect local contexts may compromise the quality of node representations. This intricate topology in web-scale graphs cannot be matched by small-scale scenarios. To address the above issues, we propose \textbf{A}daptive \textbf{T}opology-aware \textbf{P}ropagation (ATP), which reduces potential high-bias propagation and extracts structural patterns of each node in a scalable manner to improve running efficiency and predictive performance. Remarkably, ATP is crafted to be a plug-and-play node-wise propagation optimization strategy, allowing for offline execution independent of the graph learning process in a new perspective. Therefore, this approach can be seamlessly integrated into most scalable GNNs while remain orthogonal to existing node-wise propagation optimization strategies. Extensive experiments on 12 datasets, including the most representative large-scale ogbn-papers100M, have demonstrated the effectiveness of ATP. Specifically, ATP has proven to be efficient in improving the performance of prevalent scalable GNNs for semi-supervised node classification while addressing redundant computational costs.
Knowledge Graphs (KGs) have become increasingly common for representing large-scale linked data. However, their immense size has required graph learning systems to assist humans in analysis, interpretation, and pattern detection. While there have been promising results for researcher- and clinician- empowerment through a variety of KG learning systems, we identify four key deficiencies in state-of-the-art graph learning that simultaneously limit KG learning performance and diminish the ability of humans to interface optimally with these learning systems. These deficiencies are: 1) lack of expert knowledge integration, 2) instability to node degree extremity in the KG, 3) lack of consideration for uncertainty and relevance while learning, and 4) lack of explainability. Furthermore, we characterise state-of-the-art attempts to solve each of these problems and note that each attempt has largely been isolated from attempts to solve the other problems. Through a formalisation of these problems and a review of the literature that addresses them, we adopt the position that not only are deficiencies in these four key areas holding back human-KG empowerment, but that the divide-and-conquer approach to solving these problems as individual units rather than a whole is a significant barrier to the interface between humans and KG learning systems. We propose that it is only through integrated, holistic solutions to the limitations of KG learning systems that human and KG learning co-empowerment will be efficiently affected. We finally present our "Veni, Vidi, Vici" framework that sets a roadmap for effectively and efficiently shifting to a holistic co-empowerment model in both the KG learning and the broader machine learning domain.
Accounting for over 20% of the total carbon emissions, the precise estimation of on-road transportation carbon emissions is crucial for carbon emission monitoring and efficient mitigation policy formulation. However, existing estimation methods typically depend on hard-to-collect individual statistics of vehicle miles traveled to calculate emissions, thereby suffering from high data collection difficulty. To relieve this issue by utilizing the strong pattern recognition of artificial intelligence, we incorporate two sources of open data representative of the transportation demand and capacity factors, the origin-destination (OD) flow data and the road network data, to build a hierarchical heterogeneous graph learning method for on-road carbon emission estimation (HENCE). Specifically, a hierarchical graph consisting of the road network level, community level, and region level is constructed to model the multi-scale road network-based connectivity and travel connection between spatial areas. Heterogeneous graphs consisting of OD links and spatial links are further built at both the community level and region level to capture the intrinsic interactions between travel demand and road network accessibility. Extensive experiments on two large-scale real-world datasets demonstrate HENCE's effectiveness and superiority with R-squared exceeding 0.75 and outperforming baselines by 9.60% on average, validating its success in pioneering the use of artificial intelligence to empower carbon emission management and sustainability development. The implementation codes are available at this link: https://github.com/tsinghua-fib-lab/HENCE.
We present GenEFT: an effective theory framework for shedding light on the statics and dynamics of neural network generalization, and illustrate it with graph learning examples. We first investigate the generalization phase transition as data size increases, comparing experimental results with information-theory-based approximations. We find generalization in a Goldilocks zone where the decoder is neither too weak nor too powerful. We then introduce an effective theory for the dynamics of representation learning, where latent-space representations are modeled as interacting particles (repons), and find that it explains our experimentally observed phase transition between generalization and overfitting as encoder and decoder learning rates are scanned. This highlights the power of physics-inspired effective theories for bridging the gap between theoretical predictions and practice in machine learning.
Multimodal data pervades various domains, including healthcare, social media, and transportation, where multimodal graphs play a pivotal role. Machine learning on multimodal graphs, referred to as multimodal graph learning (MGL), is essential for successful artificial intelligence (AI) applications. The burgeoning research in this field encompasses diverse graph data types and modalities, learning techniques, and application scenarios. This survey paper conducts a comparative analysis of existing works in multimodal graph learning, elucidating how multimodal learning is achieved across different graph types and exploring the characteristics of prevalent learning techniques. Additionally, we delineate significant applications of multimodal graph learning and offer insights into future directions in this domain. Consequently, this paper serves as a foundational resource for researchers seeking to comprehend existing MGL techniques and their applicability across diverse scenarios.
Graph transformers typically lack third-order interactions, limiting their geometric understanding which is crucial for tasks like molecular geometry prediction. We propose the Triplet Graph Transformer (TGT) that enables direct communication between pairs within a 3-tuple of nodes via novel triplet attention and aggregation mechanisms. TGT is applied to molecular property prediction by first predicting interatomic distances from 2D graphs and then using these distances for downstream tasks. A novel three-stage training procedure and stochastic inference further improve training efficiency and model performance. Our model achieves new state-of-the-art (SOTA) results on open challenge benchmarks PCQM4Mv2 and OC20 IS2RE. We also obtain SOTA results on QM9, MOLPCBA, and LIT-PCBA molecular property prediction benchmarks via transfer learning. We also demonstrate the generality of TGT with SOTA results on the traveling salesman problem (TSP).
Invariant models, one important class of geometric deep learning models, are capable of generating meaningful geometric representations by leveraging informative geometric features in point clouds. These models are characterized by their simplicity, good experimental results and computational efficiency. However, their theoretical expressive power still remains unclear, restricting a deeper understanding of the potential of such models. In this work, we concentrate on characterizing the theoretical expressiveness of a wide range of invariant models. We first rigorously bound the expressiveness of the most classic invariant model, message-passing neural networks incorporating distance (DisGNN), restricting its unidentifiable cases to be only highly symmetric point clouds. We then show that GeoNGNN, the geometric counterpart of one of the simplest subgraph graph neural networks (subgraph GNNs), can effectively break these corner cases' symmetry and thus achieve E(3)-completeness. By leveraging GeoNGNN as a theoretical tool, we further prove that: 1) most subgraph GNNs developed in traditional graph learning can be seamlessly extended to geometric scenarios with E(3)-completeness; 2) DimeNet, GemNet and SphereNet, three well-established invariant models, are also all capable of achieving E(3)-completeness. Our theoretical results fill the gap in the theoretical power of invariant models, contributing to a rigorous and comprehensive understanding of their capabilities. We also empirically evaluated GeoNGNN, the simplest model within the large E(3)-complete family we established, which achieves competitive results to models relying on high-order invariant/equivariant representations on molecule-relevant tasks.
With the arrival of the big data era, mobility profiling has become a viable method of utilizing enormous amounts of mobility data to create an intelligent transportation system. Mobility profiling can extract potential patterns in urban traffic from mobility data and is critical for a variety of traffic-related applications. However, due to the high level of complexity and the huge amount of data, mobility profiling faces huge challenges. Digital Twin (DT) technology paves the way for cost-effective and performance-optimised management by digitally creating a virtual representation of the network to simulate its behaviour. In order to capture the complex spatio-temporal features in traffic scenario, we construct alignment diagrams to assist in completing the spatio-temporal correlation representation and design dilated alignment convolution network (DACN) to learn the fine-grained correlations, i.e., spatio-temporal interactions. We propose a digital twin mobility profiling (DTMP) framework to learn node profiles on a mobility network DT model. Extensive experiments have been conducted upon three real-world datasets. Experimental results demonstrate the effectiveness of DTMP.
Patent documents in the patent database (PatDB) are crucial for research, development, and innovation as they contain valuable technical information. However, PatDB presents a multifaceted challenge compared to publicly available preprocessed databases due to the intricate nature of the patent text and the inherent sparsity within the patent citation network. Although patent text analysis and citation analysis bring new opportunities to explore patent data mining, no existing work exploits the complementation of them. To this end, we propose a joint semantic-topological evolutionary graph learning approach (PatSTEG) to model the formation dynamics of patent citation networks. More specifically, we first create a real-world dataset of Chinese patents named CNPat and leverage its patent texts and citations to construct a patent citation network. Then, PatSTEG is modeled to study the evolutionary dynamics of patent citation formation by considering the semantic and topological information jointly. Extensive experiments are conducted on CNPat and public datasets to prove the superiority of PatSTEG over other state-of-the-art methods. All the results provide valuable references for patent literature research and technical exploration.
In this paper, we propose the first framework that enables solving graph learning tasks of all levels (node, edge and graph) and all types (generation, regression and classification) using one formulation. We first formulate prediction tasks including regression and classification into a generic (conditional) generation framework, which enables diffusion models to perform deterministic tasks with provable guarantees. We then propose Latent Graph Diffusion (LGD), a generative model that can generate node, edge, and graph-level features of all categories simultaneously. We achieve this goal by embedding the graph structures and features into a latent space leveraging a powerful encoder and decoder, then training a diffusion model in the latent space. LGD is also capable of conditional generation through a specifically designed cross-attention mechanism. Leveraging LGD and the ``all tasks as generation'' formulation, our framework is capable of solving graph tasks of various levels and types. We verify the effectiveness of our framework with extensive experiments, where our models achieve state-of-the-art or highly competitive results across a wide range of generation and regression tasks.
Graph Neural Networks (GNNs) excel in various graph learning tasks but face computational challenges when applied to large-scale graphs. A promising solution is to remove non-essential edges to reduce the computational overheads in GNN. Previous literature generally falls into two categories: topology-guided and semantic-guided. The former maintains certain graph topological properties yet often underperforms on GNNs due to low integration with neural network training. The latter performs well at lower sparsity on GNNs but faces performance collapse at higher sparsity levels. With this in mind, we take the first step to propose a new research line and concept termed Graph Sparse Training (GST), which dynamically manipulates sparsity at the data level. Specifically, GST initially constructs a topology & semantic anchor at a low training cost, followed by performing dynamic sparse training to align the sparse graph with the anchor. We introduce the Equilibria Sparsification Principle to guide this process, effectively balancing the preservation of both topological and semantic information. Ultimately, GST produces a sparse graph with maximum topological integrity and no performance degradation. Extensive experiments on 6 datasets and 5 backbones showcase that GST (I) identifies subgraphs at higher graph sparsity levels (1.67%~15.85% $\uparrow$) than state-of-the-art sparsification methods, (II) preserves more key spectral properties, (III) achieves 1.27-3.42$\times$ speedup in GNN inference and (IV) successfully helps graph adversarial defense and graph lottery tickets.
Graph-structured data, prevalent in domains ranging from social networks to biochemical analysis, serve as the foundation for diverse real-world systems. While graph neural networks demonstrate proficiency in modeling this type of data, their success is often reliant on significant amounts of labeled data, posing a challenge in practical scenarios with limited annotation resources. To tackle this problem, tremendous efforts have been devoted to enhancing graph machine learning performance under low-resource settings by exploring various approaches to minimal supervision. In this paper, we introduce a novel concept of Data-Efficient Graph Learning (DEGL) as a research frontier, and present the first survey that summarizes the current progress of DEGL. We initiate by highlighting the challenges inherent in training models with large labeled data, paving the way for our exploration into DEGL. Next, we systematically review recent advances on this topic from several key aspects, including self-supervised graph learning, semi-supervised graph learning, and few-shot graph learning. Also, we state promising directions for future research, contributing to the evolution of graph machine learning.
In Multi-Modal Knowledge Graphs (MMKGs), Multi-Modal Entity Alignment (MMEA) is crucial for identifying identical entities across diverse modal attributes. However, semantic inconsistency, mainly due to missing modal attributes, poses a significant challenge. Traditional approaches rely on attribute interpolation, but this often introduces modality noise, distorting the original semantics. Moreover, the lack of a universal theoretical framework limits advancements in achieving semantic consistency. This study introduces a novel approach, DESAlign, which addresses these issues by applying a theoretical framework based on Dirichlet energy to ensure semantic consistency. We discover that semantic inconsistency leads to model overfitting to modality noise, causing performance fluctuations, particularly when modalities are missing. DESAlign innovatively combats over-smoothing and interpolates absent semantics using existing modalities. Our approach includes a multi-modal knowledge graph learning strategy and a propagation technique that employs existing semantic features to compensate for missing ones, providing explicit Euler solutions. Comprehensive evaluations across 60 benchmark splits, including monolingual and bilingual scenarios, demonstrate that DESAlign surpasses existing methods, setting a new standard in performance. Further testing with high rates of missing modalities confirms its robustness, offering an effective solution to semantic inconsistency in real-world MMKGs.
Graph contrastive learning (GCL) has emerged as a state-of-the-art strategy for learning representations of diverse graphs including social and biomedical networks. GCL widely uses stochastic graph topology augmentation, such as uniform node dropping, to generate augmented graphs. However, such stochastic augmentations may severely damage the intrinsic properties of a graph and deteriorate the following representation learning process. We argue that incorporating an awareness of cohesive subgraphs during the graph augmentation and learning processes has the potential to enhance GCL performance. To this end, we propose a novel unified framework called CTAug, to seamlessly integrate cohesion awareness into various existing GCL mechanisms. In particular, CTAug comprises two specialized modules: topology augmentation enhancement and graph learning enhancement. The former module generates augmented graphs that carefully preserve cohesion properties, while the latter module bolsters the graph encoder's ability to discern subgraph patterns. Theoretical analysis shows that CTAug can strictly improve existing GCL mechanisms. Empirical experiments verify that CTAug can achieve state-of-the-art performance for graph representation learning, especially for graphs with high degrees. The code is available at https://doi.org/10.5281/zenodo.10594093, or https://github.com/wuyucheng2002/CTAug.
Continual learning (CL) is the research field that aims to build machine learning models that can accumulate knowledge continuously over different tasks without retraining from scratch. Previous studies have shown that pre-training graph neural networks (GNN) may lead to negative transfer (Hu et al., 2020) after fine-tuning, a setting which is closely related to CL. Thus, we focus on studying GNN in the continual graph learning (CGL) setting. We propose the first continual graph learning benchmark for spatio-temporal graphs and use it to benchmark well-known CGL methods in this novel setting. The benchmark is based on the N-UCLA and NTU-RGB+D datasets for skeleton-based action recognition. Beyond benchmarking for standard performance metrics, we study the class and task-order sensitivity of CGL methods, i.e., the impact of learning order on each class/task's performance, and the architectural sensitivity of CGL methods with backbone GNN at various widths and depths. We reveal that task-order robust methods can still be class-order sensitive and observe results that contradict previous empirical observations on architectural sensitivity in CL.
The rapidly changing landscape of technology and industries leads to dynamic skill requirements, making it crucial for employees and employers to anticipate such shifts to maintain a competitive edge in the labor market. Existing efforts in this area either rely on domain-expert knowledge or regarding skill evolution as a simplified time series forecasting problem. However, both approaches overlook the sophisticated relationships among different skills and the inner-connection between skill demand and supply variations. In this paper, we propose a Cross-view Hierarchical Graph learning Hypernetwork (CHGH) framework for joint skill demand-supply prediction. Specifically, CHGH is an encoder-decoder network consisting of i) a cross-view graph encoder to capture the interconnection between skill demand and supply, ii) a hierarchical graph encoder to model the co-evolution of skills from a cluster-wise perspective, and iii) a conditional hyper-decoder to jointly predict demand and supply variations by incorporating historical demand-supply gaps. Extensive experiments on three real-world datasets demonstrate the superiority of the proposed framework compared to seven baselines and the effectiveness of the three modules.
The transformer architecture has shown remarkable success in various domains, such as natural language processing and computer vision. When it comes to graph learning, transformers are required not only to capture the interactions between pairs of nodes but also to preserve graph structures connoting the underlying relations and proximity between them, showing the expressive power to capture different graph structures. Accordingly, various structure-preserving graph transformers have been proposed and widely used for various tasks, such as graph-level tasks in bioinformatics and chemoinformatics. However, strategies related to graph structure preservation have not been well organized and systematized in the literature. In this paper, we provide a comprehensive overview of structure-preserving graph transformers and generalize these methods from the perspective of their design objective. First, we divide strategies into four main groups: node feature modulation, context node sampling, graph rewriting, and transformer architecture improvements. We then further divide the strategies according to the coverage and goals of graph structure preservation. Furthermore, we also discuss challenges and future directions for graph transformer models to preserve the graph structure and understand the nature of graphs.