Self-Supervised Class-Cognizant Few-Shot Classification

About

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.

Ojas Kishore Shirekar, Hadi Jamali-Rad · 2022

Related benchmarks

Task	Dataset	Result
Few-shot classification	tieredImageNet (test)	Accuracy61.77	282
Few-shot Image Classification	Mini-Imagenet (test)	Accuracy—	235
Image Classification	MiniImagenet	Accuracy64.81	213
Few-shot Image Classification	tieredImageNet	Accuracy—	190
Few-shot classification	Mini-Imagenet (test)	Accuracy—	113
Few-shot classification	Omniglot (test)	Accuracy97.38	109
Few-shot classification	MiniImagenet	5-way 5-shot Accuracy64.81	98
Cross-domain few-shot classification	CD-FSL benchmark	Mean Accuracy—	33