Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions (Student Abstract)

Yuchen Xiao, Joshua Hoffman, Tian Xia, Christopher Amato

AAAI 2020 pp. 13965-13966

doi:10.1609/AAAI.V34I10.7255 /aaai/2020/xiao2020aaai-multi/

Abstract

We consider the challenges of learning multi-agent/robot macro-action-based deep Q-nets including how to properly update each macro-action value and accurately maintain macro-action-observation trajectories. We address these challenges by first proposing two fundamental frameworks for learning macro-action-value function and joint macro-action-value function. Furthermore, we present two new approaches of learning decentralized macro-action-based policies, which involve a new double Q-update rule that facilitates the learning of decentralized Q-nets by using a centralized Q-net for action selection. Our approaches are evaluated both in simulation and on real robots.

PDF AAAI Semantic Scholar

Cite

Text

Xiao et al. "Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I10.7255

Markdown

[Xiao et al. "Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/xiao2020aaai-multi/) doi:10.1609/AAAI.V34I10.7255

BibTeX

@inproceedings{xiao2020aaai-multi,
  title     = {{Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions (Student Abstract)}},
  author    = {Xiao, Yuchen and Hoffman, Joshua and Xia, Tian and Amato, Christopher},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {13965-13966},
  doi       = {10.1609/AAAI.V34I10.7255},
  url       = {https://mlanthology.org/aaai/2020/xiao2020aaai-multi/}
}