Inductive Graph Alignment Prompt: Bridging the Gap between Graph Pre-training and Inductive Fine-tuning From Spectral Perspective

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yan, Yuchen, Zhang, Peiyan, Fang, Zheng, Long, Qingqing
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911781232836608
author Yan, Yuchen
Zhang, Peiyan
Fang, Zheng
Long, Qingqing
author_facet Yan, Yuchen
Zhang, Peiyan
Fang, Zheng
Long, Qingqing
contents The "Graph pre-training and fine-tuning" paradigm has significantly improved Graph Neural Networks(GNNs) by capturing general knowledge without manual annotations for downstream tasks. However, due to the immense gap of data and tasks between the pre-training and fine-tuning stages, the model performance is still limited. Inspired by prompt fine-tuning in Natural Language Processing(NLP), many endeavors have been made to bridge the gap in graph domain. But existing methods simply reformulate the form of fine-tuning tasks to the pre-training ones. With the premise that the pre-training graphs are compatible with the fine-tuning ones, these methods typically operate in transductive setting. In order to generalize graph pre-training to inductive scenario where the fine-tuning graphs might significantly differ from pre-training ones, we propose a novel graph prompt based method called Inductive Graph Alignment Prompt(IGAP). Firstly, we unify the mainstream graph pre-training frameworks and analyze the essence of graph pre-training from graph spectral theory. Then we identify the two sources of the data gap in inductive setting: (i) graph signal gap and (ii) graph structure gap. Based on the insight of graph pre-training, we propose to bridge the graph signal gap and the graph structure gap with learnable prompts in the spectral space. A theoretical analysis ensures the effectiveness of our method. At last, we conduct extensive experiments among nodes classification and graph classification tasks under the transductive, semi-inductive and inductive settings. The results demonstrate that our proposed method can successfully bridge the data gap under different settings.
format Preprint
id arxiv_https___arxiv_org_abs_2402_13556
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Inductive Graph Alignment Prompt: Bridging the Gap between Graph Pre-training and Inductive Fine-tuning From Spectral Perspective
Yan, Yuchen
Zhang, Peiyan
Fang, Zheng
Long, Qingqing
Machine Learning
Artificial Intelligence
E.2
The "Graph pre-training and fine-tuning" paradigm has significantly improved Graph Neural Networks(GNNs) by capturing general knowledge without manual annotations for downstream tasks. However, due to the immense gap of data and tasks between the pre-training and fine-tuning stages, the model performance is still limited. Inspired by prompt fine-tuning in Natural Language Processing(NLP), many endeavors have been made to bridge the gap in graph domain. But existing methods simply reformulate the form of fine-tuning tasks to the pre-training ones. With the premise that the pre-training graphs are compatible with the fine-tuning ones, these methods typically operate in transductive setting. In order to generalize graph pre-training to inductive scenario where the fine-tuning graphs might significantly differ from pre-training ones, we propose a novel graph prompt based method called Inductive Graph Alignment Prompt(IGAP). Firstly, we unify the mainstream graph pre-training frameworks and analyze the essence of graph pre-training from graph spectral theory. Then we identify the two sources of the data gap in inductive setting: (i) graph signal gap and (ii) graph structure gap. Based on the insight of graph pre-training, we propose to bridge the graph signal gap and the graph structure gap with learnable prompts in the spectral space. A theoretical analysis ensures the effectiveness of our method. At last, we conduct extensive experiments among nodes classification and graph classification tasks under the transductive, semi-inductive and inductive settings. The results demonstrate that our proposed method can successfully bridge the data gap under different settings.
title Inductive Graph Alignment Prompt: Bridging the Gap between Graph Pre-training and Inductive Fine-tuning From Spectral Perspective
topic Machine Learning
Artificial Intelligence
E.2
url https://arxiv.org/abs/2402.13556