Class-Imbalanced Graph Learning without Class Rebalancing

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Zhining, Qiu, Ruizhong, Zeng, Zhichen, Yoo, Hyunsik, Zhou, David, Xu, Zhe, Zhu, Yada, Weldemariam, Kommy, He, Jingrui, Tong, Hanghang
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916250761822208
author Liu, Zhining
Qiu, Ruizhong
Zeng, Zhichen
Yoo, Hyunsik
Zhou, David
Xu, Zhe
Zhu, Yada
Weldemariam, Kommy
He, Jingrui
Tong, Hanghang
author_facet Liu, Zhining
Qiu, Ruizhong
Zeng, Zhichen
Yoo, Hyunsik
Zhou, David
Xu, Zhe
Zhu, Yada
Weldemariam, Kommy
He, Jingrui
Tong, Hanghang
contents Class imbalance is prevalent in real-world node classification tasks and poses great challenges for graph learning models. Most existing studies are rooted in a class-rebalancing (CR) perspective and address class imbalance with class-wise reweighting or resampling. In this work, we approach the root cause of class-imbalance bias from an topological paradigm. Specifically, we theoretically reveal two fundamental phenomena in the graph topology that greatly exacerbate the predictive bias stemming from class imbalance. On this basis, we devise a lightweight topological augmentation framework BAT to mitigate the class-imbalance bias without class rebalancing. Being orthogonal to CR, BAT can function as an efficient plug-and-play module that can be seamlessly combined with and significantly boost existing CR techniques. Systematic experiments on real-world imbalanced graph learning tasks show that BAT can deliver up to 46.27% performance gain and up to 72.74% bias reduction over existing techniques. Code, examples, and documentations are available at https://github.com/ZhiningLiu1998/BAT.
format Preprint
id arxiv_https___arxiv_org_abs_2308_14181
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Class-Imbalanced Graph Learning without Class Rebalancing
Liu, Zhining
Qiu, Ruizhong
Zeng, Zhichen
Yoo, Hyunsik
Zhou, David
Xu, Zhe
Zhu, Yada
Weldemariam, Kommy
He, Jingrui
Tong, Hanghang
Machine Learning
Artificial Intelligence
Class imbalance is prevalent in real-world node classification tasks and poses great challenges for graph learning models. Most existing studies are rooted in a class-rebalancing (CR) perspective and address class imbalance with class-wise reweighting or resampling. In this work, we approach the root cause of class-imbalance bias from an topological paradigm. Specifically, we theoretically reveal two fundamental phenomena in the graph topology that greatly exacerbate the predictive bias stemming from class imbalance. On this basis, we devise a lightweight topological augmentation framework BAT to mitigate the class-imbalance bias without class rebalancing. Being orthogonal to CR, BAT can function as an efficient plug-and-play module that can be seamlessly combined with and significantly boost existing CR techniques. Systematic experiments on real-world imbalanced graph learning tasks show that BAT can deliver up to 46.27% performance gain and up to 72.74% bias reduction over existing techniques. Code, examples, and documentations are available at https://github.com/ZhiningLiu1998/BAT.
title Class-Imbalanced Graph Learning without Class Rebalancing
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2308.14181