Learning Options with Interest Functions

AAAI 2019 pp. 9955-9956

doi:10.1609/AAAI.V33I01.33019955 /aaai/2019/khetarpal2019aaai-learning-a/

Abstract

Learning temporal abstractions which are partial solutions to a task and could be reused for solving other tasks is an ingredient that can help agents to plan and learn efficiently. In this work, we tackle this problem in the options framework. We aim to autonomously learn options which are specialized in different state space regions by proposing a notion of interest functions, which generalizes initiation sets from the options framework for function approximation. We build on the option-critic framework to derive policy gradient theorems for interest functions, leading to a new interest-option-critic architecture.

PDF AAAI Semantic Scholar

Cite

Text

Khetarpal and Precup. "Learning Options with Interest Functions." AAAI Conference on Artificial Intelligence, 2019. doi:10.1609/AAAI.V33I01.33019955

Markdown

[Khetarpal and Precup. "Learning Options with Interest Functions." AAAI Conference on Artificial Intelligence, 2019.](https://mlanthology.org/aaai/2019/khetarpal2019aaai-learning-a/) doi:10.1609/AAAI.V33I01.33019955

BibTeX

@inproceedings{khetarpal2019aaai-learning-a,
  title     = {{Learning Options with Interest Functions}},
  author    = {Khetarpal, Khimya and Precup, Doina},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {9955-9956},
  doi       = {10.1609/AAAI.V33I01.33019955},
  url       = {https://mlanthology.org/aaai/2019/khetarpal2019aaai-learning-a/}
}