BECLR: Batch Enhanced Contrastive Few-Shot Learning

Stylianos Poulakakis-Daktylidis, Hadi Jamali-Rad

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Keywords: few-shot classification, unsupervised few-shot learning, deep representation learning

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

TL;DR: We introduce BECLR , a novel end-to-end U-FSL approach to: (i) infuse implicit class-level insights within the contrastive pretraining stage and (ii) address the inherent sample bias in the (U)-FSL inference stage, achieving state-of-the-art results.

Abstract: Learning quickly from very few labeled samples is a fundamental attribute that separates machines and humans in the era of deep representation learning. Unsupervised few-shot learning (U-FSL) aspires to bridge this gap by discarding the reliance on annotations at training time. Intrigued by the success of contrastive learning approaches in the realm of U-FSL, we structurally approach their shortcomings in both pretraining and downstream inference stages. We propose a novel Dynamic Clustered mEmory (DyCE) module to promote a highly separable latent representation space for enhancing positive sampling at the pretraining phase and infusing implicit class-level insights into unsupervised contrastive learning. We then tackle the, somehow overlooked yet critical, issue of sample bias at the few-shot inference stage. We propose an iterative Optimal Transport-based distribution Alignment (OpTA) strategy and demonstrate that it efficiently addresses the problem, especially in low-shot scenarios where FSL approaches suffer the most from sample bias. We later on discuss that DyCE and OpTA are two intertwined pieces of a novel end-to-end approach (we coin as BECLR), constructively magnifying each other's impact. We then present a suite of extensive quantitative and qualitative experimentation to corroborate that BECLR sets a new state-of-the-art across ALL existing U-FSL benchmarks (to the best of our knowledge), and significantly outperforms the best of the current baselines (codebase available at https://github.com/stypoumic/BECLR).

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning

Submission Number: 1421