AlphaMatch: Improving Consistency for Semi-supervised Learning with Alpha-divergence
This work provides an incremental improvement in semi-supervised learning performance for practitioners seeking higher accuracy with limited labeled data.
This paper introduces AlphaMatch, a semi-supervised learning method that improves label consistency between data points and their augmented versions. It achieves 91.3% test accuracy on CIFAR-10 with only 4 labeled data per class, outperforming FixMatch's 88.7%.
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.