Multi-View Deep Attention Network for Reinforcement Learning (Student Abstract)

Yueyue Hu, Shiliang Sun, Xin Xu, Jing Zhao

AAAI 2020 pp. 13811-13812

doi:10.1609/AAAI.V34I10.7177 /aaai/2020/hu2020aaai-multi/

Abstract

The representation approximated by a single deep network is usually limited for reinforcement learning agents. We propose a novel multi-view deep attention network (MvDAN), which introduces multi-view representation learning into the reinforcement learning task for the first time. The proposed model approximates a set of strategies from multiple representations and combines these strategies based on attention mechanisms to provide a comprehensive strategy for a single-agent. Experimental results on eight Atari video games show that the MvDAN has effective competitive performance than single-view reinforcement learning methods.

PDF AAAI Semantic Scholar

Cite

Text

Hu et al. "Multi-View Deep Attention Network for Reinforcement Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I10.7177

Markdown

[Hu et al. "Multi-View Deep Attention Network for Reinforcement Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/hu2020aaai-multi/) doi:10.1609/AAAI.V34I10.7177

BibTeX

@inproceedings{hu2020aaai-multi,
  title     = {{Multi-View Deep Attention Network for Reinforcement Learning (Student Abstract)}},
  author    = {Hu, Yueyue and Sun, Shiliang and Xu, Xin and Zhao, Jing},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {13811-13812},
  doi       = {10.1609/AAAI.V34I10.7177},
  url       = {https://mlanthology.org/aaai/2020/hu2020aaai-multi/}
}