This is an interim version of our Electronic Legal Deposit Catalogue-eJournals and eBooks while we continue to recover from a cyber-attack.
Path integral based convolution and pooling for graph neural networks*This article is an updated version of: Ma Z, Xuan J, Wang Y G, Li M and Liò P 2020 Path integral based convolution and pooling for graph neural networks Advances in Neural Information Processing Systems vol 33 ed H Larochelle, M Ranzato, R Hadsell, M F Balcan and H Lin (New York: Curran Associates) pp 16421–33. (29th December 2021)
Record Type:
Journal Article
Title:
Path integral based convolution and pooling for graph neural networks*This article is an updated version of: Ma Z, Xuan J, Wang Y G, Li M and Liò P 2020 Path integral based convolution and pooling for graph neural networks Advances in Neural Information Processing Systems vol 33 ed H Larochelle, M Ranzato, R Hadsell, M F Balcan and H Lin (New York: Curran Associates) pp 16421–33. (29th December 2021)
Main Title:
Path integral based convolution and pooling for graph neural networks*This article is an updated version of: Ma Z, Xuan J, Wang Y G, Li M and Liò P 2020 Path integral based convolution and pooling for graph neural networks Advances in Neural Information Processing Systems vol 33 ed H Larochelle, M Ranzato, R Hadsell, M F Balcan and H Lin (New York: Curran Associates) pp 16421–33.
Abstract: Graph neural networks (GNNs) extend the functionality of traditional neural networks to graph-structured data. Similar to CNNs, an optimized design of graph convolution and pooling is key to success. Borrowing ideas from physics, we propose path integral-based GNNs (PAN) for classification and regression tasks on graphs. Specifically, we consider a convolution operation that involves every path linking the message sender and receiver with learnable weights depending on the path length, which corresponds to the maximal entropy random walk. It generalizes the graph Laplacian to a new transition matrix that we call the maximal entropy transition (MET) matrix derived from a path integral formalism. Importantly, the diagonal entries of the MET matrix are directly related to the subgraph centrality, thus leading to a natural and adaptive pooling mechanism. PAN provides a versatile framework that can be tailored for different graph data with varying sizes and structures. We can view most existing GNN architectures as special cases of PAN. Experimental results show that PAN achieves state-of-the-art performance on various graph classification/regression tasks, including a new benchmark dataset from statistical mechanics that we propose to boost applications of GNN in physical sciences.