Reinforcement Learning, Bit by Bit

Abstract

Reinforcement learning agents have demonstrated remarkable achievements in simulated environments. Data efficiency poses an impediment to carrying this success over to real environments. The design of data-efficient agents calls for a deeper understanding of information acquisition and representation. We discuss concepts and regret analysis that together offer principled guidance. This line of thinking sheds light on questions of what information to seek, how to seek that information, and what information to retain. To illustrate concepts, we design simple agents that build on them and present computational results that highlight data efficiency.

Cite

Text

Lu et al. "Reinforcement Learning, Bit by Bit." Foundations and Trends in Machine Learning, 2023. doi:10.1561/2200000097

Markdown

[Lu et al. "Reinforcement Learning, Bit by Bit." Foundations and Trends in Machine Learning, 2023.](https://mlanthology.org/ftml/2023/lu2023ftml-reinforcement/) doi:10.1561/2200000097

BibTeX

@article{lu2023ftml-reinforcement,
  title     = {{Reinforcement Learning, Bit by Bit}},
  author    = {Lu, Xiuyuan and Van Roy, Benjamin and Dwaracherla, Vikranth and Ibrahimi, Morteza and Osband, Ian and Wen, Zheng},
  journal   = {Foundations and Trends in Machine Learning},
  year      = {2023},
  pages     = {733-865},
  doi       = {10.1561/2200000097},
  volume    = {16},
  url       = {https://mlanthology.org/ftml/2023/lu2023ftml-reinforcement/}
}