Learnings Options End-to-End for Continuous Action Tasks
Machine Learning
2017-12-04 v1 Artificial Intelligence
Abstract
We present new results on learning temporally extended actions for continuoustasks, using the options framework (Suttonet al.[1999b], Precup [2000]). In orderto achieve this goal we work with the option-critic architecture (Baconet al.[2017])using a deliberation cost and train it with proximal policy optimization (Schulmanet al.[2017]) instead of vanilla policy gradient. Results on Mujoco domains arepromising, but lead to interesting questions aboutwhena given option should beused, an issue directly connected to the use of initiation sets.
Cite
@article{arxiv.1712.00004,
title = {Learnings Options End-to-End for Continuous Action Tasks},
author = {Martin Klissarov and Pierre-Luc Bacon and Jean Harb and Doina Precup},
journal= {arXiv preprint arXiv:1712.00004},
year = {2017}
}