Semi-supervised learning with max-margin graph cuts

2026-04-29

AI Summary

This work addresses the challenge of effectively leveraging limited labeled data together with abundant unlabeled data to enhance model generalization in semi-supervised learning. The authors propose a novel approach based on maximum-margin graph cuts, which integrates graph structure, harmonic functions, and the maximum-margin principle by optimizing graph cuts over pseudo-labels induced by the solution of a harmonic function. Theoretical analysis provides a corresponding generalization error bound, and empirical evaluations demonstrate that the proposed method significantly outperforms state-of-the-art baselines—including manifold-regularized support vector machines—on both synthetic data and three UCI benchmark datasets, thereby confirming its efficacy and superiority.

Abstract

This paper proposes a novel algorithm for semisupervised learning. This algorithm learns graph cuts that maximize the margin with respect to the labels induced by the harmonic function solution. We motivate the approach, compare it to existing work, and prove a bound on its generalization error. The quality of our solutions is evaluated on a synthetic problem and three UCI ML repository datasets. In most cases, we outperform manifold regularization of support vector machines, which is a state-of-the-art approach to semi-supervised max-margin learning.

Problem

• semi-supervised learning
• max-margin
• graph cuts
• generalization error
• manifold regularization

Innovation

• semi-supervised learning
• max-margin
• graph cuts
• harmonic function
• manifold regularization