Lower and upper bounds for numbers of linear regions of graph convolutional networks

DOI10.1016/J.NEUNET.2023.09.025MaRDI QIDQ6535213zbMATH OpenFDO

Authors Hao Chen, Yuguang Wang, Huan Xiong

Publication date 23 November 2023

Published in Neural Networks (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.NEUNET.2023.09.025

zbMATH Keywords

expressivityGCNsgraph convolutional networksgraph neural networkslinear regionsReLU

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07)

Recommendations

• On the number of regions of piecewise linear neural networks
• The logic of graph neural networks
• Theory of graph neural networks: representation and learning
• Understanding the message passing in graph neural networks via power iteration clustering

Cites work

• An introduction to hyperplane arrangements
• An optimal lower bound on the number of variables for graph identification
• Analytic combinatorics
• Approximation and estimation bounds for artificial neural networks
• Deep learning
• Descriptive Complexity, Canonisation, and Definable Graph Structure Theory
• Error bounds for approximations with deep ReLU networks
• Facing up to arrangements: face-count formulas for partitions of space by hyperplanes
• On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities
• PEBBLE GAMES AND LINEAR EQUATIONS
• Permitted and Forbidden Sets in Symmetric Threshold-Linear Networks
• Random Graph Isomorphism

Cited in

(3)

• On the Explainability of Graph Convolutional Network With GCN Tangent Kernel
• A Framework for the Construction of Upper Bounds on the Number of Affine Linear Regions of ReLU Feed-Forward Neural Networks
• On the number of regions of piecewise linear neural networks

This page was built for publication: Lower and upper bounds for numbers of linear regions of graph convolutional networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6535213)