Saved in:
Bibliographic Details
Main Authors: Liu, Tom, Wu, Anna, Li, Chao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.22745
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912509148004352
author Liu, Tom
Wu, Anna
Li, Chao
author_facet Liu, Tom
Wu, Anna
Li, Chao
contents Self-training has become a popular semi-supervised learning technique for leveraging unlabeled data. However, the over-confidence of pseudo-labels remains a key challenge. In this paper, we propose a novel \emph{graph-based uncertainty-aware self-training} (GUST) framework to combat over-confidence in node classification. Drawing inspiration from the uncertainty integration idea introduced by Wang \emph{et al.}~\cite{wang2024uncertainty}, our method largely diverges from previous self-training approaches by focusing on \emph{stochastic node labeling} grounded in the graph topology. Specifically, we deploy a Bayesian-inspired module to estimate node-level uncertainty, incorporate these estimates into the pseudo-label generation process via an expectation-maximization (EM)-like step, and iteratively update both node embeddings and adjacency-based transformations. Experimental results on several benchmark graph datasets demonstrate that our GUST framework achieves state-of-the-art performance, especially in settings where labeled data is extremely sparse.
format Preprint
id arxiv_https___arxiv_org_abs_2503_22745
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph-Based Uncertainty-Aware Self-Training with Stochastic Node Labeling
Liu, Tom
Wu, Anna
Li, Chao
Machine Learning
Self-training has become a popular semi-supervised learning technique for leveraging unlabeled data. However, the over-confidence of pseudo-labels remains a key challenge. In this paper, we propose a novel \emph{graph-based uncertainty-aware self-training} (GUST) framework to combat over-confidence in node classification. Drawing inspiration from the uncertainty integration idea introduced by Wang \emph{et al.}~\cite{wang2024uncertainty}, our method largely diverges from previous self-training approaches by focusing on \emph{stochastic node labeling} grounded in the graph topology. Specifically, we deploy a Bayesian-inspired module to estimate node-level uncertainty, incorporate these estimates into the pseudo-label generation process via an expectation-maximization (EM)-like step, and iteratively update both node embeddings and adjacency-based transformations. Experimental results on several benchmark graph datasets demonstrate that our GUST framework achieves state-of-the-art performance, especially in settings where labeled data is extremely sparse.
title Graph-Based Uncertainty-Aware Self-Training with Stochastic Node Labeling
topic Machine Learning
url https://arxiv.org/abs/2503.22745