Fourier Guided Adaptive Adversarial Augmentation for Generalization in Visual Reinforcement Learning
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (17) · 18110-18118
DOI: 10.1609/aaai.v39i17.33992 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Jeong Woon Lee und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
Visual Reinforcement Learning (RL) facilitates learning directly from raw images; however, the domain gap between training and testing environments frequently leads to a decline in performance within unseen environments. In this paper, we propose Fourier Guided Adaptive Adversarial Augmentation (FGA3), a novel augmentation method that maintains semantic consistency. We focus on style augmentation in the frequency domain by keeping the phase and altering the amplitude to preserve the state of the original data. For adaptive adversarial perturbation, we reformulate the worst-case problem to RL by employing adversarial example training, which leverages value loss and cosine similarity within a semantic space. Moreover, our findings illustrate that cosine similarity is effective in quantifying feature distances within a semantic space. Extensive experiments on DMControl-GB and Procgen have shown that FGA3 is compatible with a wide range of visual RL algorithms, both off-policy and on-policy, and significantly improves the robustness of the agent in unseen environments.
Zitieren
APA
Lee, J. W., & Hwang, H. (2025). Fourier Guided Adaptive Adversarial Augmentation for Generalization in Visual Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 39(17), 18110-18118. https://doi.org/10.1609/aaai.v39i17.33992
Vancouver
Lee JW, Hwang H. Fourier Guided Adaptive Adversarial Augmentation for Generalization in Visual Reinforcement Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(17):18110-18118. doi:10.1609/aaai.v39i17.33992
BibTeX
@article{Lee2025,
author = {Lee, Jeong Woon and Hwang, Hyoseok},
title = {Fourier Guided Adaptive Adversarial Augmentation for Generalization in Visual Reinforcement Learning},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {17},
pages = {18110--18118},
doi = {10.1609/aaai.v39i17.33992}
}
EUVIMED