Teaching MLP More Graph Information: A Three-stage Multitask Knowledge Distillation Framework

TL;DR

This paper introduces a three-stage multitask distillation framework that enables MLP models to effectively learn graph information, reducing reliance on graph structure and improving inference efficiency.

Contribution

It is the first to combine graph positional encoding with MLP and include hidden layer distillation for graph data, enhancing performance and generalization.

Findings

Outperforms baseline models in accuracy and robustness

Effectively captures positional information with Positional Encoding

Demonstrates stability across various experimental settings

Abstract

We study the challenging problem for inference tasks on large-scale graph datasets of Graph Neural Networks: huge time and memory consumption, and try to overcome it by reducing reliance on graph structure. Even though distilling graph knowledge to student MLP is an excellent idea, it faces two major problems of positional information loss and low generalization. To solve the problems, we propose a new three-stage multitask distillation framework. In detail, we use Positional Encoding to capture positional information. Also, we introduce Neural Heat Kernels responsible for graph data processing in GNN and utilize hidden layer outputs matching for better performance of student MLP's hidden layers. To the best of our knowledge, it is the first work to include hidden layer distillation for student MLP on graphs and to combine graph Positional Encoding with MLP. We test its performance and robustness with several settings and draw the conclusion that our work can outperform well with good stability.