Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification

Jay Gala, Pengtao Xie
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:14433-14452, 2024.

Abstract

In many image classification applications, the number of labeled training images is limited, which leads to model overfitting. To mitigate the lack of training data, deep generative models have been leveraged to generate synthetic training data. However, existing methods generate data for individual classes based on how much training data they have without considering their actual data needs. To address this limitation, we propose needs-aware image generation, which automatically identifies the different data needs of individual classes based on their classification performance and divides a limited data generation budget into these classes according to their needs. We propose a multi-level optimization based framework which performs four learning stages in an end-to-end manner. Experiments on both imbalanced and balanced classification datasets demonstrate the effectiveness of our proposed method.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-gala24a, title = {Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification}, author = {Gala, Jay and Xie, Pengtao}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {14433--14452}, 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/gala24a/gala24a.pdf}, url = {https://proceedings.mlr.press/v235/gala24a.html}, abstract = {In many image classification applications, the number of labeled training images is limited, which leads to model overfitting. To mitigate the lack of training data, deep generative models have been leveraged to generate synthetic training data. However, existing methods generate data for individual classes based on how much training data they have without considering their actual data needs. To address this limitation, we propose needs-aware image generation, which automatically identifies the different data needs of individual classes based on their classification performance and divides a limited data generation budget into these classes according to their needs. We propose a multi-level optimization based framework which performs four learning stages in an end-to-end manner. Experiments on both imbalanced and balanced classification datasets demonstrate the effectiveness of our proposed method.} }
Endnote
%0 Conference Paper %T Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification %A Jay Gala %A Pengtao Xie %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-gala24a %I PMLR %P 14433--14452 %U https://proceedings.mlr.press/v235/gala24a.html %V 235 %X In many image classification applications, the number of labeled training images is limited, which leads to model overfitting. To mitigate the lack of training data, deep generative models have been leveraged to generate synthetic training data. However, existing methods generate data for individual classes based on how much training data they have without considering their actual data needs. To address this limitation, we propose needs-aware image generation, which automatically identifies the different data needs of individual classes based on their classification performance and divides a limited data generation budget into these classes according to their needs. We propose a multi-level optimization based framework which performs four learning stages in an end-to-end manner. Experiments on both imbalanced and balanced classification datasets demonstrate the effectiveness of our proposed method.
APA
Gala, J. & Xie, P.. (2024). Leverage Class-Specific Accuracy to Guide Data Generation for Improving Image Classification. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:14433-14452 Available from https://proceedings.mlr.press/v235/gala24a.html.

Related Material