TouchUp-G: Improving Feature Representation through Graph-Centric Finetuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Jing, Song, Xiang, Ioannidis, Vassilis N., Koutra, Danai, Faloutsos, Christos
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908288806813696
author Zhu, Jing
Song, Xiang
Ioannidis, Vassilis N.
Koutra, Danai
Faloutsos, Christos
author_facet Zhu, Jing
Song, Xiang
Ioannidis, Vassilis N.
Koutra, Danai
Faloutsos, Christos
contents How can we enhance the node features acquired from Pretrained Models (PMs) to better suit downstream graph learning tasks? Graph Neural Networks (GNNs) have become the state-of-the-art approach for many high-impact, real-world graph applications. For feature-rich graphs, a prevalent practice involves utilizing a PM directly to generate features, without incorporating any domain adaptation techniques. Nevertheless, this practice is suboptimal because the node features extracted from PM are graph-agnostic and prevent GNNs from fully utilizing the potential correlations between the graph structure and node features, leading to a decline in GNNs performance. In this work, we seek to improve the node features obtained from a PM for downstream graph tasks and introduce TOUCHUP-G, which has several advantages. It is (a) General: applicable to any downstream graph task, including link prediction which is often employed in recommender systems; (b) Multi-modal: able to improve raw features of any modality (e.g. images, texts, audio); (c) Principled: it is closely related to a novel metric, feature homophily, which we propose to quantify the potential correlations between the graph structure and node features and we show that TOUCHUP-G can effectively shrink the discrepancy between the graph structure and node features; (d) Effective: achieving state-of-the-art results on four real-world datasets spanning different tasks and modalities.
format Preprint
id arxiv_https___arxiv_org_abs_2309_13885
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TouchUp-G: Improving Feature Representation through Graph-Centric Finetuning
Zhu, Jing
Song, Xiang
Ioannidis, Vassilis N.
Koutra, Danai
Faloutsos, Christos
Machine Learning
Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
Social and Information Networks
How can we enhance the node features acquired from Pretrained Models (PMs) to better suit downstream graph learning tasks? Graph Neural Networks (GNNs) have become the state-of-the-art approach for many high-impact, real-world graph applications. For feature-rich graphs, a prevalent practice involves utilizing a PM directly to generate features, without incorporating any domain adaptation techniques. Nevertheless, this practice is suboptimal because the node features extracted from PM are graph-agnostic and prevent GNNs from fully utilizing the potential correlations between the graph structure and node features, leading to a decline in GNNs performance. In this work, we seek to improve the node features obtained from a PM for downstream graph tasks and introduce TOUCHUP-G, which has several advantages. It is (a) General: applicable to any downstream graph task, including link prediction which is often employed in recommender systems; (b) Multi-modal: able to improve raw features of any modality (e.g. images, texts, audio); (c) Principled: it is closely related to a novel metric, feature homophily, which we propose to quantify the potential correlations between the graph structure and node features and we show that TOUCHUP-G can effectively shrink the discrepancy between the graph structure and node features; (d) Effective: achieving state-of-the-art results on four real-world datasets spanning different tasks and modalities.
title TouchUp-G: Improving Feature Representation through Graph-Centric Finetuning
topic Machine Learning
Artificial Intelligence
Computation and Language
Computer Vision and Pattern Recognition
Social and Information Networks
url https://arxiv.org/abs/2309.13885