ICML 2016poster805 citations
Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning
Abstract
In this paper we present a new way of predicting the performance of a reinforcement learning policy given historical data that may have been generated by a different policy. The ability to evaluate a policy from historical data is important for applications where the deployment of a bad policy can be dangerous or costly. We show empirically that our algorithm produces estimates that often have orders of magnitude lower mean squared error than existing methods—it makes more efficient use of the available data. Our new estimator is based on two advances: an extension of the doubly robust estimator (Jiang & Li, 2015), and a new way to mix between model based and importance sampling based estimates.
BibTeX
@InProceedings{pmlr-v48-thomasa16,
title = {Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning},
author = {Thomas, Philip and Brunskill, Emma},
booktitle = {Proceedings of The 33rd International Conference on Machine Learning},
pages = {2139--2148},
year = {2016},
editor = {Balcan, Maria Florina and Weinberger, Kilian Q.},
volume = {48},
series = {Proceedings of Machine Learning Research},
address = {New York, New York, USA},
month = {20--22 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v48/thomasa16.pdf},
url = {https://proceedings.mlr.press/v48/thomasa16.html},
abstract = {In this paper we present a new way of predicting the performance of a reinforcement learning policy given historical data that may have been generated by a different policy. The ability to evaluate a policy from historical data is important for applications where the deployment of a bad policy can be dangerous or costly. We show empirically that our algorithm produces estimates that often have orders of magnitude lower mean squared error than existing methods—it makes more efficient use of the available data. Our new estimator is based on two advances: an extension of the doubly robust estimator (Jiang & Li, 2015), and a new way to mix between model based and importance sampling based estimates.}
}