VQGraph: Rethinking Graph Representation Space for Bridging GNNs and MLPs

TL;DR

VQGraph introduces a novel graph representation space using a structure-aware tokenizer with VQ-VAE, enabling more effective GNN-to-MLP knowledge distillation and achieving state-of-the-art results.

Contribution

The paper proposes a new graph representation space via a VQ-VAE based tokenizer, improving GNN-to-MLP distillation by capturing local structural information.

Findings

Achieves state-of-the-art GNN-to-MLP distillation performance across seven datasets.

VQGraph infers 828x faster than GNNs while maintaining high accuracy.

Improves MLP accuracy by 28.05% over standalone MLPs.

Abstract

GNN-to-MLP distillation aims to utilize knowledge distillation (KD) to learn computationally-efficient multi-layer perceptron (student MLP) on graph data by mimicking the output representations of teacher GNN. Existing methods mainly make the MLP to mimic the GNN predictions over a few class labels. However, the class space may not be expressive enough for covering numerous diverse local graph structures, thus limiting the performance of knowledge transfer from GNN to MLP. To address this issue, we propose to learn a new powerful graph representation space by directly labeling nodes' diverse local structures for GNN-to-MLP distillation. Specifically, we propose a variant of VQ-VAE to learn a structure-aware tokenizer on graph data that can encode each node's local substructure as a discrete code. The discrete codes constitute a codebook as a new graph representation space that is able to identify different local graph structures of nodes with the corresponding code indices. Then, based on the learned codebook, we propose a new distillation target, namely soft code assignments, to directly transfer the structural knowledge of each node from GNN to MLP. The resulting framework VQGraph achieves new state-of-the-art performance on GNN-to-MLP distillation in both transductive and inductive settings across seven graph datasets. We show that VQGraph with better performance infers faster than GNNs by 828x, and also achieves accuracy improvement over GNNs and stand-alone MLPs by 3.90% and 28.05% on average, respectively. Code: https://github.com/YangLing0818/VQGraph.