Graph Convolutional Networks with EigenPooling

Ma, Yao; Wang, Suhang; Aggarwal, Charų C.; Tang, Jiliang

doi:10.1145/3292500.3330982

Cited by 285 publications

(142 citation statements)

References 41 publications

Supporting

Mentioning

139

Contrasting

Order By: Relevance

“…Recently, graph convolutional network (GCN) models [4,9,17,23,33,35] have been widely studied. For example, [3] first designs the graph convolution operation in Fourier domain by the graph Laplacian.…”

Section: Related Workmentioning

confidence: 99%

AM-GCN: Adaptive Multi-channel Graph Convolutional Networks

Wang

Zhu

et al. 2020

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

397

136

View full text Add to dashboard Cite

Graph Convolutional Networks (GCNs) have gained great popularity in tackling various analytics tasks on graph and network data. However, some recent studies raise concerns about whether GCNs can optimally integrate node features and topological structures in a complex graph with rich information. In this paper, we first present an experimental investigation. Surprisingly, our experimental results clearly show that the capability of the state-of-the-art GCNs in fusing node features and topological structures is distant from optimal or even satisfactory. The weakness may severely hinder the capability of GCNs in some classification tasks, since GCNs may not be able to adaptively learn some deep correlation information between topological structures and node features. Can we remedy the weakness and design a new type of GCNs that can retain the advantages of the state-of-the-art GCNs and, at the same time, enhance the capability of fusing topological structures and node features substantially? We tackle the challenge and propose an adaptive multi-channel graph convolutional networks for semi-supervised classification (AM-GCN). The central idea is that we extract the specific and common embeddings from node features, topological structures, and their combinations simultaneously, and use the attention mechanism to learn adaptive importance weights of the embeddings. Our extensive experiments on benchmark data sets clearly show that AM-GCN extracts the most correlated information from both node features and topological structures substantially, and improves the classification accuracy with a clear margin.

show abstract

Section: Related Workmentioning

confidence: 99%

AM-GCN: Adaptive Multi-channel Graph Convolutional Networks

Wang

Zhu

et al. 2020

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

397

136

View full text Add to dashboard Cite

show abstract

“…It is used to build hierarchical GNNs, where hierarchical graph pooling is used several times between GNN layers to gradually decrease the number of nodes. The most representative hierarchical graph pooling methods are DIFFPOOL [17], SORTPOOL [16], TOPKPOOL [18], SAG-POOL [19], and EIGENPOOL [20]. A straightforward way to use hierarchical graph pooling for graph representation learning is to reduce the number of nodes to one.…”

Section: Graph Pooling: Global Versus Hierarchicalmentioning

confidence: 99%

“…As discussed in Section 2.1, they are used in flat GNNs after all GNN layers and output the graph representation for the classifier. While flat GNNs outperform hierarchical GNNs in most benchmark datasets [12], developing hierarchical graph pooling is still desired, especially for large graphs [17], [19], [20]. Therefore, we explore a hierarchical graph pooling method based on second-order pooling.…”

Section: Multi-head Attentional Second-order Poolingmentioning

confidence: 99%

“…However, averaging and summation ignore the feature correlation information, hampering the overall model performance [16]. Other studies have proposed advanced graph pooling methods, including DIFFPOOL [17], SORT-POOL [16], TOPKPOOL [18], SAGPOOL [19], and EIGEN-POOL [20]. DIFFPOOL maps nodes to a pre-defined number of clusters but is hard to train.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Second-Order Pooling for Graph Neural Networks

Wang

2023

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

Graph neural networks have achieved great success in learning node representations for graph tasks such as node classification and link prediction. Graph representation learning requires graph pooling to obtain graph representations from node representations. It is challenging to develop graph pooling methods due to the variable sizes and isomorphic structures of graphs. In this work, we propose to use second-order pooling as graph pooling, which naturally solves the above challenges. In addition, compared to existing graph pooling methods, second-order pooling is able to use information from all nodes and collect second-order statistics, making it more powerful. We show that direct use of second-order pooling with graph neural networks leads to practical problems. To overcome these problems, we propose two novel global graph pooling methods based on second-order pooling; namely, bilinear mapping and attentional second-order pooling. In addition, we extend attentional second-order pooling to hierarchical graph pooling for more flexible use in GNNs. We perform thorough experiments on graph classification tasks to demonstrate the effectiveness and superiority of our proposed methods. Experimental results show that our methods improve the performance significantly and consistently.

show abstract

“…Inspired by the remarkable development of graph neural networks in various domains [13] [40] [27] [39], researchers have noticed the potentials of them for molecular property prediction. Generally, by treating the molecule as a graph, several graph neural networks have been applied [14,24,40] as an architecture that can directly deal with noneuclidean data like graphs. Variants of graph neural networks like MPNN [13], Schnet [30], can be applied for molecular properties prediction where they use nodes to represent atoms, and the edges are weighted by the distances between atoms.…”

Section: Related Workmentioning

confidence: 99%

ASGN: An Active Semi-supervised Graph Neural Network for Molecular Property Prediction

Hao

Huang

et al. 2020

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery &Amp; Data Mining

101

View full text Add to dashboard Cite

Molecular property prediction (e.g., energy) is an essential problem in chemistry and biology. Unfortunately, many supervised learning methods usually suffer from the problem of scarce labeled molecules in the chemical space, where such property labels are generally obtained by Density Functional Theory (DFT) calculation which is extremely computational costly. An effective solution is to incorporate the unlabeled molecules in a semi-supervised fashion. However, learning semi-supervised representation for large amounts of molecules is challenging, including the joint representation issue of both molecular essence and structure, the conflict between representation and property leaning. Here we propose a novel framework called Active Semi-supervised Graph Neural Network (ASGN) by incorporating both labeled and unlabeled molecules. Specifically, ASGN adopts a teacher-student framework. In the teacher model, we propose a novel semi-supervised learning method to learn general representation that jointly exploits information from molecular structure and molecular distribution. Then in the student model, we target at property prediction task to deal with the learning loss conflict. At last, we proposed a novel active learning strategy in terms of molecular diversities to select informative data during the whole framework learning. We conduct extensive experiments on several public datasets. Experimental results show the remarkable performance of our ASGN framework.

show abstract

Graph Convolutional Networks with EigenPooling

Cited by 285 publications

References 41 publications

AM-GCN: Adaptive Multi-channel Graph Convolutional Networks

AM-GCN: Adaptive Multi-channel Graph Convolutional Networks

Second-Order Pooling for Graph Neural Networks

ASGN: An Active Semi-supervised Graph Neural Network for Molecular Property Prediction

Contact Info

Product

Resources

About