Self-Supervised Class-Cognizant Few-Shot Classification

TL;DR

This paper introduces a self-supervised, class-aware contrastive learning approach that significantly improves few-shot classification performance on standard and cross-domain benchmarks by incorporating class-level information during pre-training.

Contribution

It extends contrastive learning with class-level cognizance through clustering and re-ranking, achieving state-of-the-art results in few-shot classification tasks.

Findings

Sets new state-of-the-art in mini-ImageNet 5-way 1 and 5-shot tasks.

Achieves top performance in cross-domain CDFSL benchmarks.

Demonstrates effectiveness of class-aware contrastive pre-training.

Abstract

Unsupervised learning is argued to be the dark matter of human intelligence. To build in this direction, this paper focuses on unsupervised learning from an abundance of unlabeled data followed by few-shot fine-tuning on a downstream classification task. To this aim, we extend a recent study on adopting contrastive learning for self-supervised pre-training by incorporating class-level cognizance through iterative clustering and re-ranking and by expanding the contrastive optimization loss to account for it. To our knowledge, our experimentation both in standard and cross-domain scenarios demonstrate that we set a new state-of-the-art (SoTA) in (5-way, 1 and 5-shot) settings of standard mini-ImageNet benchmark as well as the (5-way, 5 and 20-shot) settings of cross-domain CDFSL benchmark. Our code and experimentation can be found in our GitHub repository: https://github.com/ojss/c3lr.