Can Increasing Input Dimensionality Improve Deep Reinforcement Learning?

Kei Ota, Tomoaki Oiki, Devesh Jha, Toshisada Mariyama, Daniel Nikovski
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:7424-7433, 2020.

Abstract

Deep reinforcement learning (RL) algorithms have recently achieved remarkable successes in various sequential decision making tasks, leveraging advances in methods for training large deep networks. However, these methods usually require large amounts of training data, which is often a big problem for real-world applications. One natural question to ask is whether learning good representations for states and using larger networks helps in learning better policies. In this paper, we try to study if increasing input dimensionality helps improve performance and sample efficiency of model-free deep RL algorithms. To do so, we propose an online feature extractor network (OFENet) that uses neural nets to produce \emph{good} representations to be used as inputs to an off-policy RL algorithm. Even though the high dimensionality of input is usually thought to make learning of RL agents more difficult, we show that the RL agents in fact learn more efficiently with the high-dimensional representation than with the lower-dimensional state observations. We believe that stronger feature propagation together with larger networks allows RL agents to learn more complex functions of states and thus improves the sample efficiency. Through numerical experiments, we show that the proposed method achieves much higher sample efficiency and better performance. Codes for the proposed method are available at http://www.merl.com/research/license/OFENet

Cite this Paper

BibTeX
@InProceedings{pmlr-v119-ota20a, title = {Can Increasing Input Dimensionality Improve Deep Reinforcement Learning?}, author = {Ota, Kei and Oiki, Tomoaki and Jha, Devesh and Mariyama, Toshisada and Nikovski, Daniel}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {7424--7433}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/ota20a/ota20a.pdf}, url = {https://proceedings.mlr.press/v119/ota20a.html}, abstract = {Deep reinforcement learning (RL) algorithms have recently achieved remarkable successes in various sequential decision making tasks, leveraging advances in methods for training large deep networks. However, these methods usually require large amounts of training data, which is often a big problem for real-world applications. One natural question to ask is whether learning good representations for states and using larger networks helps in learning better policies. In this paper, we try to study if increasing input dimensionality helps improve performance and sample efficiency of model-free deep RL algorithms. To do so, we propose an online feature extractor network (OFENet) that uses neural nets to produce \emph{good} representations to be used as inputs to an off-policy RL algorithm. Even though the high dimensionality of input is usually thought to make learning of RL agents more difficult, we show that the RL agents in fact learn more efficiently with the high-dimensional representation than with the lower-dimensional state observations. We believe that stronger feature propagation together with larger networks allows RL agents to learn more complex functions of states and thus improves the sample efficiency. Through numerical experiments, we show that the proposed method achieves much higher sample efficiency and better performance. Codes for the proposed method are available at http://www.merl.com/research/license/OFENet} }
Endnote
%0 Conference Paper %T Can Increasing Input Dimensionality Improve Deep Reinforcement Learning? %A Kei Ota %A Tomoaki Oiki %A Devesh Jha %A Toshisada Mariyama %A Daniel Nikovski %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-ota20a %I PMLR %P 7424--7433 %U https://proceedings.mlr.press/v119/ota20a.html %V 119 %X Deep reinforcement learning (RL) algorithms have recently achieved remarkable successes in various sequential decision making tasks, leveraging advances in methods for training large deep networks. However, these methods usually require large amounts of training data, which is often a big problem for real-world applications. One natural question to ask is whether learning good representations for states and using larger networks helps in learning better policies. In this paper, we try to study if increasing input dimensionality helps improve performance and sample efficiency of model-free deep RL algorithms. To do so, we propose an online feature extractor network (OFENet) that uses neural nets to produce \emph{good} representations to be used as inputs to an off-policy RL algorithm. Even though the high dimensionality of input is usually thought to make learning of RL agents more difficult, we show that the RL agents in fact learn more efficiently with the high-dimensional representation than with the lower-dimensional state observations. We believe that stronger feature propagation together with larger networks allows RL agents to learn more complex functions of states and thus improves the sample efficiency. Through numerical experiments, we show that the proposed method achieves much higher sample efficiency and better performance. Codes for the proposed method are available at http://www.merl.com/research/license/OFENet
APA
Ota, K., Oiki, T., Jha, D., Mariyama, T. & Nikovski, D.. (2020). Can Increasing Input Dimensionality Improve Deep Reinforcement Learning?. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:7424-7433 Available from https://proceedings.mlr.press/v119/ota20a.html.

Related Material