Coarsely-Labeled Data for Better Few-Shot Transfer

ICCV 2021 pp. 9052-9061

doi:10.1109/ICCV48922.2021.00892 /iccv/2021/phoo2021iccv-coarselylabeled/

Abstract

Few-shot learning is based on the premise that labels are expensive, especially when they are fine-grained and require expertise. But coarse labels might be easy to acquire and thus abundant. We present a representation learning approach - PAS that allows few-shot learners to leverage coarsely-labeled data available before evaluation. Inspired by self-training, we label the additional data using a teacher trained on the base dataset and filter the teacher's prediction based on the coarse labels; a new student representation is then trained on the base dataset and the pseudo-labeled dataset. PAS is able to produce a representation that consistently and significantly outperforms the baselines in 3 different datasets. Code is available at https://github.com/cpphoo/PAS.

PDF ICCV Semantic Scholar

Cite

Text

Phoo and Hariharan. "Coarsely-Labeled Data for Better Few-Shot Transfer." International Conference on Computer Vision, 2021. doi:10.1109/ICCV48922.2021.00892

Markdown

[Phoo and Hariharan. "Coarsely-Labeled Data for Better Few-Shot Transfer." International Conference on Computer Vision, 2021.](https://mlanthology.org/iccv/2021/phoo2021iccv-coarselylabeled/) doi:10.1109/ICCV48922.2021.00892

BibTeX

@inproceedings{phoo2021iccv-coarselylabeled,
  title     = {{Coarsely-Labeled Data for Better Few-Shot Transfer}},
  author    = {Phoo, Cheng Perng and Hariharan, Bharath},
  booktitle = {International Conference on Computer Vision},
  year      = {2021},
  pages     = {9052-9061},
  doi       = {10.1109/ICCV48922.2021.00892},
  url       = {https://mlanthology.org/iccv/2021/phoo2021iccv-coarselylabeled/}
}