Spatial Contrastive Learning for Few-Shot Classification

Abstract

In this paper, we explore contrastive learning for few-shot classification, in which we propose to use it as an additional auxiliary training objective acting as a data-dependent regularizer to promote more general and transferable features. In particular, we present a novel attention-based spatial contrastive objective to learn locally discriminative and class-agnostic features. As a result, our approach overcomes some of the limitations of the cross-entropy loss, such as its excessive discrimination towards seen classes, which reduces the transferability of features to unseen classes. With extensive experiments, we show that the proposed method outperforms state-of-the-art approaches, confirming the importance of learning good and transferable embeddings for few-shot learning.

Cite

Text

Ouali et al. "Spatial Contrastive Learning for Few-Shot Classification." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021. doi:10.1007/978-3-030-86486-6_41

Markdown

[Ouali et al. "Spatial Contrastive Learning for Few-Shot Classification." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021.](https://mlanthology.org/ecmlpkdd/2021/ouali2021ecmlpkdd-spatial/) doi:10.1007/978-3-030-86486-6_41

BibTeX

@inproceedings{ouali2021ecmlpkdd-spatial,
  title     = {{Spatial Contrastive Learning for Few-Shot Classification}},
  author    = {Ouali, Yassine and Hudelot, Céline and Tami, Myriam},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2021},
  pages     = {671-686},
  doi       = {10.1007/978-3-030-86486-6_41},
  url       = {https://mlanthology.org/ecmlpkdd/2021/ouali2021ecmlpkdd-spatial/}
}