MaxMatch: Semi-Supervised Learning with Worst-Case Consistency

TL;DR

This paper introduces MaxMatch, a semi-supervised learning method that employs worst-case consistency regularization, providing theoretical insights and demonstrating improved performance on benchmark datasets.

Contribution

It proposes a novel worst-case consistency regularization technique with theoretical guarantees, bridging the gap between theory and practice in SSL.

Findings

Effective on five benchmark datasets

Theoretically converges to a stationary point

Outperforms existing SSL methods

Abstract

In recent years, great progress has been made to incorporate unlabeled data to overcome the inefficiently supervised problem via semi-supervised learning (SSL). Most state-of-the-art models are based on the idea of pursuing consistent model predictions over unlabeled data toward the input noise, which is called consistency regularization. Nonetheless, there is a lack of theoretical insights into the reason behind its success. To bridge the gap between theoretical and practical results, we propose a worst-case consistency regularization technique for SSL in this paper. Specifically, we first present a generalization bound for SSL consisting of the empirical loss terms observed on labeled and unlabeled training data separately. Motivated by this bound, we derive an SSL objective that minimizes the largest inconsistency between an original unlabeled sample and its multiple augmented variants. We then provide a simple but effective algorithm to solve the proposed minimax problem, and theoretically prove that it converges to a stationary point. Experiments on five popular benchmark datasets validate the effectiveness of our proposed method.