General Policy Mapping: Online Continual Reinforcement Learning Inspired on the Insect Brain

Abstract

We have developed a model for online continual reinforcement learning (RL) inspired on the insect brain. Our model leverages the offline training of a feature extraction and a common general policy layer to enable the convergence of RL algorithms in online settings. Sharing a common policy layer across tasks leads to positive backward transfer, where the agent continuously improved in older tasks sharing the same underlying general policy. Biologically inspired restrictions to the agent's network are key for the convergence of RL algorithms. This provides a pathway towards efficient online RL in resource-constrained scenarios.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

Yanguas-Gil and Madireddy. "General Policy Mapping: Online Continual Reinforcement Learning Inspired on the Insect Brain." NeurIPS 2022 Workshops: Offline_RL, 2022.

Markdown

[Yanguas-Gil and Madireddy. "General Policy Mapping: Online Continual Reinforcement Learning Inspired on the Insect Brain." NeurIPS 2022 Workshops: Offline_RL, 2022.](https://mlanthology.org/neuripsw/2022/yanguasgil2022neuripsw-general/)

BibTeX

@inproceedings{yanguasgil2022neuripsw-general,
  title     = {{General Policy Mapping: Online Continual Reinforcement Learning Inspired on the Insect Brain}},
  author    = {Yanguas-Gil, Angel and Madireddy, Sandeep},
  booktitle = {NeurIPS 2022 Workshops: Offline_RL},
  year      = {2022},
  url       = {https://mlanthology.org/neuripsw/2022/yanguasgil2022neuripsw-general/}
}