Multi-Agent Pattern Formation with Deep Reinforcement Learning (Student Abstract)

Elhadji Amadou Oury Diallo, Toshiharu Sugawara

AAAI 2020 pp. 13779-13780

doi:10.1609/AAAI.V34I10.7161 /aaai/2020/diallo2020aaai-multi/

Abstract

We propose a decentralized multi-agent deep reinforcement learning architecture to investigate pattern formation under the local information provided by the agents' sensors. It consists of tasking a large number of homogeneous agents to move to a set of specified goal locations, addressing both the assignment and trajectory planning sub-problems concurrently. We then show that agents trained on random patterns can organize themselves into very complex shapes.

PDF AAAI Semantic Scholar

Cite

Text

Diallo and Sugawara. "Multi-Agent Pattern Formation with Deep Reinforcement Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I10.7161

Markdown

[Diallo and Sugawara. "Multi-Agent Pattern Formation with Deep Reinforcement Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/diallo2020aaai-multi/) doi:10.1609/AAAI.V34I10.7161

BibTeX

@inproceedings{diallo2020aaai-multi,
  title     = {{Multi-Agent Pattern Formation with Deep Reinforcement Learning (Student Abstract)}},
  author    = {Diallo, Elhadji Amadou Oury and Sugawara, Toshiharu},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {13779-13780},
  doi       = {10.1609/AAAI.V34I10.7161},
  url       = {https://mlanthology.org/aaai/2020/diallo2020aaai-multi/}
}