Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yu, Haiyang, Lee, Meng-Chieh, song, Xiang, Zhu, Qi, Faloutsos, Christos
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2512.06236
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908707001991168
author Yu, Haiyang
Lee, Meng-Chieh
song, Xiang
Zhu, Qi
Faloutsos, Christos
author_facet Yu, Haiyang
Lee, Meng-Chieh
song, Xiang
Zhu, Qi
Faloutsos, Christos
contents We explore the node classification task in the context of graph domain adaptation, which uses both source and target graph structures along with source labels to enhance the generalization capabilities of Graph Neural Networks (GNNs) on target graphs. Structure domain shifts frequently occur, especially when graph data are collected at different times or from varying areas, resulting in poor performance of GNNs on target graphs. Surprisingly, we find that simply incorporating an auxiliary loss function for denoising graph edges on target graphs can be extremely effective in enhancing GNN performance on target graphs. Based on this insight, we propose our framework, GraphDeT, a framework that integrates this auxiliary edge task into GNN training for node classification under domain adaptation. Our theoretical analysis connects this auxiliary edge task to the graph generalization bound with -distance, demonstrating such auxiliary task can imposes a constraint which tightens the bound and thereby improves generalization. The experimental results demonstrate superior performance compared to the existing baselines in handling both time and regional domain graph shifts.
format Preprint
id arxiv_https___arxiv_org_abs_2512_06236
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Empowering GNNs for Domain Adaptation via Denoising Target Graph
Yu, Haiyang
Lee, Meng-Chieh
song, Xiang
Zhu, Qi
Faloutsos, Christos
Machine Learning
We explore the node classification task in the context of graph domain adaptation, which uses both source and target graph structures along with source labels to enhance the generalization capabilities of Graph Neural Networks (GNNs) on target graphs. Structure domain shifts frequently occur, especially when graph data are collected at different times or from varying areas, resulting in poor performance of GNNs on target graphs. Surprisingly, we find that simply incorporating an auxiliary loss function for denoising graph edges on target graphs can be extremely effective in enhancing GNN performance on target graphs. Based on this insight, we propose our framework, GraphDeT, a framework that integrates this auxiliary edge task into GNN training for node classification under domain adaptation. Our theoretical analysis connects this auxiliary edge task to the graph generalization bound with -distance, demonstrating such auxiliary task can imposes a constraint which tightens the bound and thereby improves generalization. The experimental results demonstrate superior performance compared to the existing baselines in handling both time and regional domain graph shifts.
title Empowering GNNs for Domain Adaptation via Denoising Target Graph
topic Machine Learning
url https://arxiv.org/abs/2512.06236