On Graph Neural Networks versus Graph-Augmented MLPs

Abstract

From the perspective of expressive power, this work compares multi-layer Graph Neural Networks (GNNs) with a simplified alternative that we call Graph-Augmented Multi-Layer Perceptrons (GA-MLPs), which first augments node features with certain multi-hop operators on the graph and then applies an MLP in a node-wise fashion. From the perspective of graph isomorphism testing, we show both theoretically and numerically that GA-MLPs with suitable operators can distinguish almost all non-isomorphic graphs, just like the Weifeiler-Lehman (WL) test. However, by viewing them as node-level functions and examining the equivalence classes they induce on rooted graphs, we prove a separation in expressive power between GA-MLPs and GNNs that grows exponentially in depth. In particular, unlike GNNs, GA-MLPs are unable to count the number of attributed walks. We also demonstrate via community detection experiments that GA-MLPs can be limited by their choice of operator family, as compared to GNNs with higher flexibility in learning.

References in corpus (21)

• Adam: A Method for Stochastic Optimization
• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• Geometric deep learning: going beyond Euclidean data
• Instance Normalization: The Missing Ingredient for Fast Stylization
• Spectral Networks and Locally Connected Networks on Graphs
• Simplifying Graph Convolutional Networks
• Spectral redemption: clustering sparse networks
• Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting
• DeeperGCN: All You Need to Train Deeper GCNs
• Revisiting Graph Neural Networks: All We Have is Low-Pass Filters
• SIGN: Scalable Inception Graph Neural Networks
• Supervised Community Detection with Line Graph Neural Networks
• Distance Encoding: Design Provably More Powerful Neural Networks for Graph Representation Learning
• On the Bottleneck of Graph Neural Networks and its Practical Implications
• Position-aware Graph Neural Networks
• A Survey on The Expressive Power of Graph Neural Networks
• Generalization and Representational Limits of Graph Neural Networks
• Are Powerful Graph Neural Nets Necessary? A Dissection on Graph Classification
• Can Graph Neural Networks Count Substructures?
• Understanding Isomorphism Bias in Graph Data Sets

Cited by in corpus (2)

• Graph-less Neural Networks: Teaching Old MLPs New Tricks via Distillation
• Towards Quantized Model Parallelism for Graph-Augmented MLPs Based on Gradient-Free ADMM Framework