Graph Neural Networks for Graphs with Heterophily: A Survey

Zheng, Xiaohui; Liu, Yixin; Pan, Shirui; Zhang, Miao; Jin, Dayong; Yu, Philip S.

doi:10.48550/arxiv.2202.07082

Cited by 17 publications

(23 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the characteristics of the data, the study considers only graph regression. [42,31,38,40] have been conducted. They classify existing GNNs into several categories, e.g., recurrent graph neural networks, convolutional graph networks, graph autoencoders, and spatial-temporal graph neural networks in [31].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs

Maekawa¹,

Noda²,

Sasaki³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph Neural Networks (GNNs) have achieved great success on a node classification task. Despite the broad interest in developing and evaluating GNNs, they have been assessed with limited benchmark datasets. As a result, the existing evaluation of GNNs lacks fine-grained analysis from various characteristics of graphs. Motivated by this, we conduct extensive experiments with a synthetic graph generator that can generate graphs having controlled characteristics for fine-grained analysis. Our empirical studies clarify the strengths and weaknesses of GNNs from four major characteristics of real-world graphs with class labels of nodes, i.e., 1) class size distributions (balanced vs. imbalanced), 2) edge connection proportions between classes (homophilic vs. heterophilic), 3) attribute values (biased vs. random), and 4) graph sizes (small vs. large). In addition, to foster future research on GNNs, we publicly release our codebase that allows users to evaluate various GNNs with various graphs. We hope this work offers interesting insights for future research.Preprint. Under review.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Also, [10] provide the recent overview of GNNs. A survey [40] has addressed summarizing the current state of GNNs for graphs with the heterophily property. Since their purpose is to summarize a line of research on GNNs and discuss potential research directions, they provide no experimental results.…”

Section: Related Workmentioning

confidence: 99%

Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs

Maekawa¹,

Noda²,

Sasaki³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Various heterogeneous graph neural networks (HGNNs) have been proposed to capture semantic information, achieving great performance in heterogeneous graph representation learning [1,26,31,39]. HGNNs are at the heart of a broad range of applications such as social network analysis [33,40], recommendation systems [3,38], and knowledge graph inference [14,18,36].…”

Section: Introductionmentioning

confidence: 99%

Simple and Efficient Heterogeneous Graph Neural Network

Yang¹,

Yan²,

Pan³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Heterogeneous graph neural networks (HGNNs) deliver the powerful capability to embed rich structural and semantic information of a heterogeneous graph into low-dimensional node representations. Existing HGNNs usually learn to embed information using hierarchy attention mechanism and repeated neighbor aggregation, suffering from unnecessary complexity and redundant computation. This paper proposes Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN) which reduces this excess complexity through avoiding overused node-level attention within the same relation and pre-computing the neighbor aggregation in the pre-processing stage. Unlike previous work, SeHGNN utilizes a light-weight parameter-free neighbor aggregator to learn structural information for each metapath, and a transformer-based semantic aggregator to combine semantic information across metapaths for the final embedding of each node. As a result, SeHGNN offers the 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 the accuracy and training speed. Codes are available at https://github.com/ICT-GIMLab/SeHGNN.

show abstract

“…Equipped with initial residues, the final representation implicitly leverages features of all fused levels. However, despite its ability to eliminate over-smoothing, models with initial residual connections in fact employees the homophily assumption [26], that is, the assumption that connected nodes tend to share similar features and labels, which might be unsuitable for heterophilic graphs [29,47]. Some variants are also explored [24,45], but they still fall into the scope of homophily.…”

Section: Introductionmentioning

confidence: 99%

Clenshaw Graph Neural Networks

Guo¹,

Wei²

2022

Preprint

View full text Add to dashboard Cite

Graph Convolutional Networks (GCNs), which use a messagepassing paradigm with stacked convolution layers, are foundational methods for learning graph representations. Recent GCN models use various residual connection techniques to alleviate the model degradation problem such as over-smoothing and gradient vanishing. Existing residual connection techniques, however, fail to make extensive use of underlying graph structure as in the graph spectral domain, which is critical for obtaining satisfactory results on heterophilic graphs.In this paper, we introduce ClenshawGCN, a GNN model that employs the Clenshaw Summation Algorithm to enhance the expressiveness of the GCN model. ClenshawGCN equips the standard GCN model with two straightforward residual modules: the adaptive initial residual connection and the negative second-order residual connection. We show that by adding these two residual modules, ClenshawGCN implicitly simulates a polynomial filter under the Chebyshev basis, giving it at least as much expressive power as polynomial spectral GNNs. In addition, we conduct comprehensive experiments to demonstrate the superiority of our model over spatial and spectral GNN models.

show abstract

Graph Neural Networks for Graphs with Heterophily: A Survey

Cited by 17 publications

References 4 publications

Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs

Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs

Simple and Efficient Heterogeneous Graph Neural Network

Clenshaw Graph Neural Networks

Contact Info

Product

Resources

About