Max-Margin Contrastive Learning

TL;DR

Max-Margin Contrastive Learning (MMCL) improves unsupervised visual representation learning by selecting support vector-based negatives to maximize decision margins, leading to faster convergence and superior performance.

Contribution

This paper introduces MMCL, a novel contrastive learning method inspired by SVMs that selects negatives as support vectors to enhance learning efficiency and effectiveness.

Findings

Outperforms state-of-the-art on standard vision benchmarks

Exhibits faster empirical convergence

Achieves better unsupervised representation quality

Abstract

Standard contrastive learning approaches usually require a large number of negatives for effective unsupervised learning and often exhibit slow convergence. We suspect this behavior is due to the suboptimal selection of negatives used for offering contrast to the positives. We counter this difficulty by taking inspiration from support vector machines (SVMs) to present max-margin contrastive learning (MMCL). Our approach selects negatives as the sparse support vectors obtained via a quadratic optimization problem, and contrastiveness is enforced by maximizing the decision margin. As SVM optimization can be computationally demanding, especially in an end-to-end setting, we present simplifications that alleviate the computational burden. We validate our approach on standard vision benchmark datasets, demonstrating better performance in unsupervised representation learning over state-of-the-art, while having better empirical convergence properties.