Policy Shaping with Human Teachers
Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence. Special Track on Machine Learning. Pages 3366-3371.
Abstract
In this work we evaluate the performance of a policy shaping algorithm using 26 human teachers. We examine if the algorithm is suitable for human-generated data on two different boards in a pac-man domain, comparing performance to an oracle that provides critique based on one known winning policy. Perhaps surprisingly, we show that the data generated by our 26 participants yields even better performance for the agent than data generated by the oracle. This might be because humans do not discourage exploring multiple winning policies. Additionally, we evaluate the impact of different verbal instructions, and different interpretations of silence, finding that the usefulness of data is affected both by what instructions is given to teachers, and how the data is interpreted.
@inproceedings{ijcai2015p474,
title = {Policy Shaping with Human Teachers},
author = {Cederborg, Thomas and Grover, Ishaan and Jr., Charles Lee Isbell and Thomaz, Andrea Lockerd},
booktitle = {Proceedings of the Twenty-Fourth International Joint Conference on
Artificial Intelligence, {IJCAI-15}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Qiang Yang and Michael Wooldridge},
pages = {3366--3371},
year = {2015},
month = {7},
note = {Special Track on Machine Learning},
url = {https://ijcai.org/proceedings/2015/474},
}
TY - CPAPER TI - Policy Shaping with Human Teachers AU - Cederborg, Thomas AU - Grover, Ishaan AU - Jr., Charles Lee Isbell AU - Thomaz, Andrea Lockerd T2 - Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-15 PB - International Joint Conferences on Artificial Intelligence PY - 2015 SP - 3366 EP - 3371 UR - https://ijcai.org/proceedings/2015/474 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
