Rangwani, H and Mopuri, KR and Babu, RV (2021) Class Balancing GAN with a Classifier in the Loop. In: 37th Conference on Uncertainty in Artificial Intelligence, UAI 2021, 27 - 30 July 2021, pp. 1618-1627.
PDF
uai_2021.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy |
Abstract
Generative Adversarial Networks (GANs) have swiftly evolved to imitate increasingly complex image distributions. However, majority of the developments focus on performance of GANs on balanced datasets. We find that the existing GANs and their training regimes which work well on balanced datasets fail to be effective in case of imbalanced (i.e. long-tailed) datasets. In this work we introduce a novel theoretically motivated Class Balancing regularizer for training GANs. Our regularizer makes use of the knowledge from a pretrained classifier to ensure balanced learning of all the classes in the dataset. This is achieved via modelling the effective class frequency based on the exponential forgetting observed in neural networks and encouraging the GAN to focus on underrepresented classes. We demonstrate the utility of our regularizer in learning representations for long-tailed distributions via achieving better performance than existing approaches over multiple datasets. Specifically, when applied to an unconditional GAN, it improves the FID from 13.03 to 9.01 on the long-tailed iNaturalist-2019 dataset.
Item Type: | Conference Paper |
---|---|
Publication: | Proceedings of Machine Learning Research |
Publisher: | ML Research Press |
Additional Information: | The copyright for this article belongs to the Association For Uncertainty in Artificial Intelligence (AUAI). |
Keywords: | Balancing; Classification (of information), Balanced datasets; Balanced learning; Complex image; Exponential forgetting; Image distributions; Long-tailed distributions; Neural-networks; Performance; Regularizer; Via modeling, Generative adversarial networks |
Department/Centre: | Division of Interdisciplinary Sciences > Computational and Data Sciences |
Date Deposited: | 25 Jul 2023 10:47 |
Last Modified: | 25 Jul 2023 10:47 |
URI: | https://eprints.iisc.ac.in/id/eprint/82662 |
Actions (login required)
View Item |