Recent developments in Transformers have opened new interesting areas of research in partially observable reinforcement learning tasks. Results from late 2019 showed that Transformers are able to outperform LSTMs on both memory intense and reactive tasks. In this work we first partially replicate the results shown in Stabilizing Transformers in RL on both reactive and memory based environments. We then show performance improvement coupled with reduced computation when adding adaptive attention span to this Stable Transformer on a challenging DMLab30 environment. The code for all our experiments and models is available at https://github.com/jerrodparker20/adaptive-transformers-in-rl.
@article{arxiv.2004.03761,
title = {Adaptive Transformers in RL},
author = {Shakti Kumar and Jerrod Parker and Panteha Naderian},
journal= {arXiv preprint arXiv:2004.03761},
year = {2020}
}
Comments
10 pages with 9 figures and 4 tables. Main text is 6 pages, appendix is 4 pages