MICo: Improved Representations via Sampling-Based State Similarity for Markov Decision Processes
Abstract
We present a new behavioural distance over the state space of a Markov decision process, and demonstrate the use of this distance as an effective means of shaping the learnt representations of deep reinforcement learning agents. While existing notions of state similarity are typically difficult to learn at scale due to high computational cost and lack of sample-based algorithms, our newly-proposed distance addresses both of these issues. In addition to providing detailed theoretical analyses, we provide empirical evidence that learning this distance alongside the value function yields structured and informative representations, including strong results on the Arcade Learning Environment benchmark.
Cite
Text
Castro et al. "MICo: Improved Representations via Sampling-Based State Similarity for Markov Decision Processes." Neural Information Processing Systems, 2021.Markdown
[Castro et al. "MICo: Improved Representations via Sampling-Based State Similarity for Markov Decision Processes." Neural Information Processing Systems, 2021.](https://mlanthology.org/neurips/2021/castro2021neurips-mico/)BibTeX
@inproceedings{castro2021neurips-mico,
title = {{MICo: Improved Representations via Sampling-Based State Similarity for Markov Decision Processes}},
author = {Castro, Pablo Samuel and Kastner, Tyler and Panangaden, Prakash and Rowland, Mark},
booktitle = {Neural Information Processing Systems},
year = {2021},
url = {https://mlanthology.org/neurips/2021/castro2021neurips-mico/}
}