A Teacher-Free Graph Knowledge Distillation Framework With Dual Self-Distillation
- 20 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 36 (9) , 4375-4385
Abstract
Recent years have witnessed great success in handling graph-related tasks with Graph Neural Networks (GNNs). Despite their great academic success, Multi-Layer Perceptrons (MLPs) remain the primary workhorse for practical industrial applications. One reason for such an academic-industry gap is the neighborhood-fetching latency incurred by data dependency in GNNs. To reduce their gaps, Graph Knowledge Distillation (GKD) is proposed, usually based on a standard teacher-student architecture, to distill knowledge from a large teacher GNN into a lightweight student GNN or MLP. However, we found in this paper that neither teachers nor GNNs are necessary for graph knowledge distillation. We propose a Teacher-Free Graph Self-Distillation (TGS) framework that does not require any teacher model or GNNs during both training and inference. More importantly, the proposed TGS framework is purely based on MLPs, where structural information is only implicitly used to guide dual knowledge self-distillation between the target node and its neighborhood. As a result, TGS enjoys the benefits of graph topology awareness in training but is free from data dependency in inference. Extensive experiments have shown that the performance of vanilla MLPs can be greatly improved with dual self-distillation, e.g., TGS improves over vanilla MLPs by 15.54% on average and outperforms state-of-the-art GKD algorithms on six real-world datasets. In terms of inference speed, TGS infers 75×-89× faster than existing GNNs and 16×-25× faster than classical inference acceleration methods.Keywords
All Related Versions
Funding Information
- National Science and Technology Major Project (2022ZD0115101)
- National Natural Science Foundation of China (U21A20427, WU2022A009)
- Westlake University
This publication has 37 references indexed in Scilit:
- HiSTGNN: Hierarchical spatio-temporal graph neural network for weather forecastingInformation Sciences, 2023
- Graph neural network for traffic forecasting: A surveyExpert Systems with Applications, 2022
- Self-Supervised Learning on Graphs: Contrastive, Generative, or PredictiveIEEE Transactions on Knowledge and Data Engineering, 2021
- Variational Graph Neural Networks for Road Traffic Prediction in Intelligent Transportation SystemsIEEE Transactions on Industrial Informatics, 2020
- Distilling Knowledge From Graph Convolutional Networks2020
- Deep Learning on Graphs: A SurveyIEEE Transactions on Knowledge and Data Engineering, 2020
- Graph neural networks: A review of methods and applicationsAI Open, 2020
- Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation2019
- Label Propagation for Deep Semi-Supervised Learning2019
- Learning Distilled Graph for Large-Scale Social Network Data ClusteringIEEE Transactions on Knowledge and Data Engineering, 2019