Fine-grained Classes and How to Find Them

Matej Grcic, Artyom Gadetsky, Maria Brbic
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:16275-16294, 2024.

Abstract

In many practical applications, coarse-grained labels are readily available compared to fine-grained labels that reflect subtle differences between classes. However, existing methods cannot leverage coarse labels to infer fine-grained labels in an unsupervised manner. To bridge this gap, we propose FALCON, a method that discovers fine-grained classes from coarsely labeled data without any supervision at the fine-grained level. FALCON simultaneously infers unknown fine-grained classes and underlying relationships between coarse and fine-grained classes. Moreover, FALCON is a modular method that can effectively learn from multiple datasets labeled with different strategies. We evaluate FALCON on eight image classification tasks and a single-cell classification task. FALCON outperforms baselines by a large margin, achieving 22% improvement over the best baseline on the tieredImageNet dataset with over 600 fine-grained classes.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-grcic24a, title = {Fine-grained Classes and How to Find Them}, author = {Grcic, Matej and Gadetsky, Artyom and Brbic, Maria}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {16275--16294}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/grcic24a/grcic24a.pdf}, url = {https://proceedings.mlr.press/v235/grcic24a.html}, abstract = {In many practical applications, coarse-grained labels are readily available compared to fine-grained labels that reflect subtle differences between classes. However, existing methods cannot leverage coarse labels to infer fine-grained labels in an unsupervised manner. To bridge this gap, we propose FALCON, a method that discovers fine-grained classes from coarsely labeled data without any supervision at the fine-grained level. FALCON simultaneously infers unknown fine-grained classes and underlying relationships between coarse and fine-grained classes. Moreover, FALCON is a modular method that can effectively learn from multiple datasets labeled with different strategies. We evaluate FALCON on eight image classification tasks and a single-cell classification task. FALCON outperforms baselines by a large margin, achieving 22% improvement over the best baseline on the tieredImageNet dataset with over 600 fine-grained classes.} }
Endnote
%0 Conference Paper %T Fine-grained Classes and How to Find Them %A Matej Grcic %A Artyom Gadetsky %A Maria Brbic %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-grcic24a %I PMLR %P 16275--16294 %U https://proceedings.mlr.press/v235/grcic24a.html %V 235 %X In many practical applications, coarse-grained labels are readily available compared to fine-grained labels that reflect subtle differences between classes. However, existing methods cannot leverage coarse labels to infer fine-grained labels in an unsupervised manner. To bridge this gap, we propose FALCON, a method that discovers fine-grained classes from coarsely labeled data without any supervision at the fine-grained level. FALCON simultaneously infers unknown fine-grained classes and underlying relationships between coarse and fine-grained classes. Moreover, FALCON is a modular method that can effectively learn from multiple datasets labeled with different strategies. We evaluate FALCON on eight image classification tasks and a single-cell classification task. FALCON outperforms baselines by a large margin, achieving 22% improvement over the best baseline on the tieredImageNet dataset with over 600 fine-grained classes.
APA
Grcic, M., Gadetsky, A. & Brbic, M.. (2024). Fine-grained Classes and How to Find Them. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:16275-16294 Available from https://proceedings.mlr.press/v235/grcic24a.html.

Related Material