AlphaMatch: Improving Consistency for Semi-supervised Learning with Alpha-divergence

TL;DR

AlphaMatch introduces a flexible alpha-divergence-based regularization and an EM-like optimization algorithm to enhance semi-supervised learning, achieving superior accuracy on standard benchmarks with fewer labeled data.

Contribution

It proposes a novel SSL method using alpha-divergence for confidence-based regularization and an EM-like algorithm for better convergence, outperforming prior methods.

Findings

Achieves 91.3% accuracy on CIFAR-10 with minimal labels

Outperforms FixMatch and other SSL methods on benchmarks

Provides a simple, effective, and easy-to-implement SSL approach

Abstract

Semi-supervised learning (SSL) is a key approach toward more data-efficient machine learning by jointly leverage both labeled and unlabeled data. We propose AlphaMatch, an efficient SSL method that leverages data augmentations, by efficiently enforcing the label consistency between the data points and the augmented data derived from them. Our key technical contribution lies on: 1) using alpha-divergence to prioritize the regularization on data with high confidence, achieving a similar effect as FixMatch but in a more flexible fashion, and 2) proposing an optimization-based, EM-like algorithm to enforce the consistency, which enjoys better convergence than iterative regularization procedures used in recent SSL methods such as FixMatch, UDA, and MixMatch. AlphaMatch is simple and easy to implement, and consistently outperforms prior arts on standard benchmarks, e.g. CIFAR-10, SVHN, CIFAR-100, STL-10. Specifically, we achieve 91.3% test accuracy on CIFAR-10 with just 4 labelled data per class, substantially improving over the previously best 88.7% accuracy achieved by FixMatch.