Back to Search Start Over

Flatness Improves Backbone Generalisation in Few-shot Classification

Authors :
Li, Rui
Trapp, Martin
Klasson, Marcus
Solin, Arno
Publication Year :
2024

Abstract

Deployment of deep neural networks in real-world settings typically requires adaptation to new tasks with few examples. Few-shot classification (FSC) provides a solution to this problem by leveraging pre-trained backbones for fast adaptation to new classes. Surprisingly, most efforts have only focused on developing architectures for easing the adaptation to the target domain without considering the importance of backbone training for good generalisation. We show that flatness-aware backbone training with vanilla fine-tuning results in a simpler yet competitive baseline compared to the state-of-the-art. Our results indicate that for in- and cross-domain FSC, backbone training is crucial to achieving good generalisation across different adaptation methods. We advocate more care should be taken when training these models.

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2404.07696
Document Type :
Working Paper