Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics
Abstract
The overestimation bias is one of the major impediments to accurate off-policy learning. This paper investigates a novel way to alleviate the overestimation bias in a continuous control setting. Our method—Truncated Quantile Critics, TQC,—blends three ideas: distributional representation of a critic, truncation of critics prediction, and ensembling of multiple critics. Distributional representation and truncation allow for arbitrary granular overestimation control, while ensembling provides additional score improvements. TQC outperforms the current state of the art on all environments from the continuous control benchmark suite, demonstrating 25% improvement on the most challenging Humanoid environment.
Cite
Text
Kuznetsov et al. "Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics." International Conference on Machine Learning, 2020.Markdown
[Kuznetsov et al. "Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics." International Conference on Machine Learning, 2020.](https://mlanthology.org/icml/2020/kuznetsov2020icml-controlling/)BibTeX
@inproceedings{kuznetsov2020icml-controlling,
title = {{Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics}},
author = {Kuznetsov, Arsenii and Shvechikov, Pavel and Grishin, Alexander and Vetrov, Dmitry},
booktitle = {International Conference on Machine Learning},
year = {2020},
pages = {5556-5566},
volume = {119},
url = {https://mlanthology.org/icml/2020/kuznetsov2020icml-controlling/}
}