Rethinking the Discount Factor in Reinforcement Learning: A Decision Theoretic Approach

AAAI 2019 pp. 7949-7956

doi:10.1609/AAAI.V33I01.33017949 /aaai/2019/pitis2019aaai-rethinking/

Abstract

Reinforcement learning (RL) agents have traditionally been tasked with maximizing the value function of a Markov decision process (MDP), either in continuous settings, with fixed discount factor γ

PDF AAAI Semantic Scholar

Cite

Text

Pitis. "Rethinking the Discount Factor in Reinforcement Learning: A Decision Theoretic Approach." AAAI Conference on Artificial Intelligence, 2019. doi:10.1609/AAAI.V33I01.33017949

Markdown

[Pitis. "Rethinking the Discount Factor in Reinforcement Learning: A Decision Theoretic Approach." AAAI Conference on Artificial Intelligence, 2019.](https://mlanthology.org/aaai/2019/pitis2019aaai-rethinking/) doi:10.1609/AAAI.V33I01.33017949

BibTeX

@inproceedings{pitis2019aaai-rethinking,
  title     = {{Rethinking the Discount Factor in Reinforcement Learning: A Decision Theoretic Approach}},
  author    = {Pitis, Silviu},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {7949-7956},
  doi       = {10.1609/AAAI.V33I01.33017949},
  url       = {https://mlanthology.org/aaai/2019/pitis2019aaai-rethinking/}
}