Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning
Abstract
To improve the sample efficiency of vision-based deep reinforcement learning (RL), we propose a novel method, called SPIRL, to automatically extract important patches from input images. Following Masked Auto-Encoders, SPIRL is based on Vision Transformer models pre-trained in a self-supervised fashion to reconstruct images from randomly-sampled patches. These pre-trained models can then be exploited to detect and select salient patches, defined as hard to reconstruct from neighboring patches. In RL, the SPIRL agent processes selected salient patches via an attention module. We empirically validate SPIRL on Atari games to test its data-efficiency against relevant state-of-the-art methods, including some traditional model-based methods and keypoint-based models. In addition, we analyze our model's interpretability capabilities.
Cite
Text
Jiang and Weng. "Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2023. doi:10.1007/978-3-031-43421-1_33Markdown
[Jiang and Weng. "Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2023.](https://mlanthology.org/ecmlpkdd/2023/jiang2023ecmlpkdd-unsupervised-a/) doi:10.1007/978-3-031-43421-1_33BibTeX
@inproceedings{jiang2023ecmlpkdd-unsupervised-a,
title = {{Unsupervised Salient Patch Selection for Data-Efficient Reinforcement Learning}},
author = {Jiang, Zhaohui and Weng, Paul},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2023},
pages = {556-572},
doi = {10.1007/978-3-031-43421-1_33},
url = {https://mlanthology.org/ecmlpkdd/2023/jiang2023ecmlpkdd-unsupervised-a/}
}