SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning

Kimin Lee, Michael Laskin, Aravind Srinivas, Pieter Abbeel
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:6131-6141, 2021.

Abstract

Off-policy deep reinforcement learning (RL) has been successful in a range of challenging domains. However, standard off-policy RL algorithms can suffer from several issues, such as instability in Q-learning and balancing exploration and exploitation. To mitigate these issues, we present SUNRISE, a simple unified ensemble method, which is compatible with various off-policy RL algorithms. SUNRISE integrates two key ingredients: (a) ensemble-based weighted Bellman backups, which re-weight target Q-values based on uncertainty estimates from a Q-ensemble, and (b) an inference method that selects actions using the highest upper-confidence bounds for efficient exploration. By enforcing the diversity between agents using Bootstrap with random initialization, we show that these different ideas are largely orthogonal and can be fruitfully integrated, together further improving the performance of existing off-policy RL algorithms, such as Soft Actor-Critic and Rainbow DQN, for both continuous and discrete control tasks on both low-dimensional and high-dimensional environments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-lee21g, title = {SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning}, author = {Lee, Kimin and Laskin, Michael and Srinivas, Aravind and Abbeel, Pieter}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {6131--6141}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/lee21g/lee21g.pdf}, url = {https://proceedings.mlr.press/v139/lee21g.html}, abstract = {Off-policy deep reinforcement learning (RL) has been successful in a range of challenging domains. However, standard off-policy RL algorithms can suffer from several issues, such as instability in Q-learning and balancing exploration and exploitation. To mitigate these issues, we present SUNRISE, a simple unified ensemble method, which is compatible with various off-policy RL algorithms. SUNRISE integrates two key ingredients: (a) ensemble-based weighted Bellman backups, which re-weight target Q-values based on uncertainty estimates from a Q-ensemble, and (b) an inference method that selects actions using the highest upper-confidence bounds for efficient exploration. By enforcing the diversity between agents using Bootstrap with random initialization, we show that these different ideas are largely orthogonal and can be fruitfully integrated, together further improving the performance of existing off-policy RL algorithms, such as Soft Actor-Critic and Rainbow DQN, for both continuous and discrete control tasks on both low-dimensional and high-dimensional environments.} }
Endnote
%0 Conference Paper %T SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning %A Kimin Lee %A Michael Laskin %A Aravind Srinivas %A Pieter Abbeel %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-lee21g %I PMLR %P 6131--6141 %U https://proceedings.mlr.press/v139/lee21g.html %V 139 %X Off-policy deep reinforcement learning (RL) has been successful in a range of challenging domains. However, standard off-policy RL algorithms can suffer from several issues, such as instability in Q-learning and balancing exploration and exploitation. To mitigate these issues, we present SUNRISE, a simple unified ensemble method, which is compatible with various off-policy RL algorithms. SUNRISE integrates two key ingredients: (a) ensemble-based weighted Bellman backups, which re-weight target Q-values based on uncertainty estimates from a Q-ensemble, and (b) an inference method that selects actions using the highest upper-confidence bounds for efficient exploration. By enforcing the diversity between agents using Bootstrap with random initialization, we show that these different ideas are largely orthogonal and can be fruitfully integrated, together further improving the performance of existing off-policy RL algorithms, such as Soft Actor-Critic and Rainbow DQN, for both continuous and discrete control tasks on both low-dimensional and high-dimensional environments.
APA
Lee, K., Laskin, M., Srinivas, A. & Abbeel, P.. (2021). SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:6131-6141 Available from https://proceedings.mlr.press/v139/lee21g.html.

Related Material