Incremental Learning of Planning Actions in Model-Based Reinforcement Learning

Abstract

The soundness and optimality of a plan depends on the correctness of the domain model. Specifying complete domain models can be difficult when interactions between an agent and its environment are complex. We propose a model-based reinforcement learning (MBRL) approach to solve planning problems with unknown models. The model is learned incrementally over episodes using only experiences from the current episode which suits non-stationary environments. We introduce the novel concept of reliability as an intrinsic motivation for MBRL, and a method to learn from failure to prevent repeated instances of similar failures. Our motivation is to improve the learning efficiency and goal-directedness of MBRL. We evaluate our work with experimental results for three planning domains.

Cite

Text

Ng and Petrick. "Incremental Learning of Planning Actions in Model-Based Reinforcement Learning." International Joint Conference on Artificial Intelligence, 2019. doi:10.24963/IJCAI.2019/443

Markdown

[Ng and Petrick. "Incremental Learning of Planning Actions in Model-Based Reinforcement Learning." International Joint Conference on Artificial Intelligence, 2019.](https://mlanthology.org/ijcai/2019/ng2019ijcai-incremental/) doi:10.24963/IJCAI.2019/443

BibTeX

@inproceedings{ng2019ijcai-incremental,
  title     = {{Incremental Learning of Planning Actions in Model-Based Reinforcement Learning}},
  author    = {Ng, Jun Hao Alvin and Petrick, Ronald P. A.},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {3195-3201},
  doi       = {10.24963/IJCAI.2019/443},
  url       = {https://mlanthology.org/ijcai/2019/ng2019ijcai-incremental/}
}