Towards a Simple Approach to Multi-step Model-based Reinforcement Learning

Kavosh Asadi; Evan Cater; Dipendra Misra; Michael L. Littman

Towards a Simple Approach to Multi-step Model-based Reinforcement Learning

Machine Learning 2018-11-02 v1 Artificial Intelligence Machine Learning

Authors: Kavosh Asadi , Evan Cater , Dipendra Misra , Michael L. Littman

Abstract

When environmental interaction is expensive, model-based reinforcement learning offers a solution by planning ahead and avoiding costly mistakes. Model-based agents typically learn a single-step transition model. In this paper, we propose a multi-step model that predicts the outcome of an action sequence with variable length. We show that this model is easy to learn, and that the model can make policy-conditional predictions. We report preliminary results that show a clear advantage for the multi-step model compared to its one-step counterpart.

Keywords

machine learning theory imitation learning reinforcement learning

Cite

@article{arxiv.1811.00128,
  title  = {Towards a Simple Approach to Multi-step Model-based Reinforcement Learning},
  author = {Kavosh Asadi and Evan Cater and Dipendra Misra and Michael L. Littman},
  journal= {arXiv preprint arXiv:1811.00128},
  year   = {2018}
}

Related papers

View all related →