Adversarial oversampling for multi-class imbalanced data classification with convolutional neural networks

Adam Wojciechowski, Mateusz Lango
Proceedings of the Fourth International Workshop on Learning with Imbalanced Domains: Theory and Applications, PMLR 183:98-111, 2022.

Abstract

Although many methods have been proposed for dealing with class imbalance, the problem of multi-class imbalanced classification still received significantly smaller attention. This problem is particularly important in image imbalanced classification since it has many critical applications, e.g., in the medical domain. One group of effective methods for imbalanced data are oversampling algorithms; however, they are usually not designed to work with image data. The current methods also work in separation from the learning algorithm, not considering the difficulties encountered during the training. In this work, we propose a new oversampling algorithm for neural networks that changes oversampled instances during training to further expand the decision region of minority classes, providing better recognition of minority classes. Experiments performed on various datasets with several configurations of class-imbalanced distributions demonstrate that the proposed method provides significant F-measure and G-mean improvements on imbalanced classification tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v183-wojciechowski22a, title = {Adversarial oversampling for multi-class imbalanced data classification with convolutional neural networks}, author = {Wojciechowski, Adam and Lango, Mateusz}, booktitle = {Proceedings of the Fourth International Workshop on Learning with Imbalanced Domains: Theory and Applications}, pages = {98--111}, year = {2022}, editor = {Moniz, Nuno and Branco, Paula and Torgo, Luís and Japkowicz, Nathalie and Wozniak, Michal and Wang, Shuo}, volume = {183}, series = {Proceedings of Machine Learning Research}, month = {23 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v183/wojciechowski22a/wojciechowski22a.pdf}, url = {https://proceedings.mlr.press/v183/wojciechowski22a.html}, abstract = {Although many methods have been proposed for dealing with class imbalance, the problem of multi-class imbalanced classification still received significantly smaller attention. This problem is particularly important in image imbalanced classification since it has many critical applications, e.g., in the medical domain. One group of effective methods for imbalanced data are oversampling algorithms; however, they are usually not designed to work with image data. The current methods also work in separation from the learning algorithm, not considering the difficulties encountered during the training. In this work, we propose a new oversampling algorithm for neural networks that changes oversampled instances during training to further expand the decision region of minority classes, providing better recognition of minority classes. Experiments performed on various datasets with several configurations of class-imbalanced distributions demonstrate that the proposed method provides significant F-measure and G-mean improvements on imbalanced classification tasks.} }
Endnote
%0 Conference Paper %T Adversarial oversampling for multi-class imbalanced data classification with convolutional neural networks %A Adam Wojciechowski %A Mateusz Lango %B Proceedings of the Fourth International Workshop on Learning with Imbalanced Domains: Theory and Applications %C Proceedings of Machine Learning Research %D 2022 %E Nuno Moniz %E Paula Branco %E Luís Torgo %E Nathalie Japkowicz %E Michal Wozniak %E Shuo Wang %F pmlr-v183-wojciechowski22a %I PMLR %P 98--111 %U https://proceedings.mlr.press/v183/wojciechowski22a.html %V 183 %X Although many methods have been proposed for dealing with class imbalance, the problem of multi-class imbalanced classification still received significantly smaller attention. This problem is particularly important in image imbalanced classification since it has many critical applications, e.g., in the medical domain. One group of effective methods for imbalanced data are oversampling algorithms; however, they are usually not designed to work with image data. The current methods also work in separation from the learning algorithm, not considering the difficulties encountered during the training. In this work, we propose a new oversampling algorithm for neural networks that changes oversampled instances during training to further expand the decision region of minority classes, providing better recognition of minority classes. Experiments performed on various datasets with several configurations of class-imbalanced distributions demonstrate that the proposed method provides significant F-measure and G-mean improvements on imbalanced classification tasks.
APA
Wojciechowski, A. & Lango, M.. (2022). Adversarial oversampling for multi-class imbalanced data classification with convolutional neural networks. Proceedings of the Fourth International Workshop on Learning with Imbalanced Domains: Theory and Applications, in Proceedings of Machine Learning Research 183:98-111 Available from https://proceedings.mlr.press/v183/wojciechowski22a.html.

Related Material