Saved in:
Bibliographic Details
Main Authors: Lei, Pui Ieng, Chen, Ximing, Sheng, Yijun, Liu, Yanyan, Gong, Zhiguo, Yang, Qiang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.17443
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914561955725312
author Lei, Pui Ieng
Chen, Ximing
Sheng, Yijun
Liu, Yanyan
Gong, Zhiguo
Yang, Qiang
author_facet Lei, Pui Ieng
Chen, Ximing
Sheng, Yijun
Liu, Yanyan
Gong, Zhiguo
Yang, Qiang
contents Existing machine learning literature lacks graph-based domain adaptation techniques capable of handling large distribution shifts, primarily due to the difficulty in simulating a coherent evolutionary path from source to target graph. To meet this challenge, we present a graph gradual domain adaptation (GGDA) framework, which constructs a compact domain sequence that minimizes information loss during adaptation. Our approach starts with an efficient generation of knowledge-preserving intermediate graphs over the Fused Gromov-Wasserstein (FGW) metric. A GGDA domain sequence is then constructed upon this bridging data pool through a novel vertex-based progression, which involves selecting "close" vertices and performing adaptive domain advancement to enhance inter-domain transferability. Theoretically, our framework provides implementable upper and lower bounds for the intractable inter-domain Wasserstein distance, $W_p(μ_t,μ_{t+1})$, enabling its flexible adjustment for optimal domain formation. Extensive experiments across diverse transfer scenarios demonstrate the superior performance of our GGDA framework.
format Preprint
id arxiv_https___arxiv_org_abs_2501_17443
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Gradual Domain Adaptation for Graph Learning
Lei, Pui Ieng
Chen, Ximing
Sheng, Yijun
Liu, Yanyan
Gong, Zhiguo
Yang, Qiang
Machine Learning
Existing machine learning literature lacks graph-based domain adaptation techniques capable of handling large distribution shifts, primarily due to the difficulty in simulating a coherent evolutionary path from source to target graph. To meet this challenge, we present a graph gradual domain adaptation (GGDA) framework, which constructs a compact domain sequence that minimizes information loss during adaptation. Our approach starts with an efficient generation of knowledge-preserving intermediate graphs over the Fused Gromov-Wasserstein (FGW) metric. A GGDA domain sequence is then constructed upon this bridging data pool through a novel vertex-based progression, which involves selecting "close" vertices and performing adaptive domain advancement to enhance inter-domain transferability. Theoretically, our framework provides implementable upper and lower bounds for the intractable inter-domain Wasserstein distance, $W_p(μ_t,μ_{t+1})$, enabling its flexible adjustment for optimal domain formation. Extensive experiments across diverse transfer scenarios demonstrate the superior performance of our GGDA framework.
title Gradual Domain Adaptation for Graph Learning
topic Machine Learning
url https://arxiv.org/abs/2501.17443