Quality-Diversity Actor-Critic: Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics

Luca Grillotti, Maxence Faldor, Borja G. León, Antoine Cully
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:16416-16459, 2024.

Abstract

A key aspect of intelligence is the ability to demonstrate a broad spectrum of behaviors for adapting to unexpected situations. Over the past decade, advancements in deep reinforcement learning have led to groundbreaking achievements to solve complex continuous control tasks. However, most approaches return only one solution specialized for a specific problem. We introduce Quality-Diversity Actor-Critic (QDAC), an off-policy actor-critic deep reinforcement learning algorithm that leverages a value function critic and a successor features critic to learn high-performing and diverse behaviors. In this framework, the actor optimizes an objective that seamlessly unifies both critics using constrained optimization to (1) maximize return, while (2) executing diverse skills. Compared with other Quality-Diversity methods, QDAC achieves significantly higher performance and more diverse behaviors on six challenging continuous control locomotion tasks. We also demonstrate that we can harness the learned skills to adapt better than other baselines to five perturbed environments. Finally, qualitative analyses showcase a range of remarkable behaviors: adaptive-intelligent-robotics.github.io/QDAC.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-grillotti24a, title = {Quality-Diversity Actor-Critic: Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics}, author = {Grillotti, Luca and Faldor, Maxence and G. Le\'{o}n, Borja and Cully, Antoine}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {16416--16459}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/grillotti24a/grillotti24a.pdf}, url = {https://proceedings.mlr.press/v235/grillotti24a.html}, abstract = {A key aspect of intelligence is the ability to demonstrate a broad spectrum of behaviors for adapting to unexpected situations. Over the past decade, advancements in deep reinforcement learning have led to groundbreaking achievements to solve complex continuous control tasks. However, most approaches return only one solution specialized for a specific problem. We introduce Quality-Diversity Actor-Critic (QDAC), an off-policy actor-critic deep reinforcement learning algorithm that leverages a value function critic and a successor features critic to learn high-performing and diverse behaviors. In this framework, the actor optimizes an objective that seamlessly unifies both critics using constrained optimization to (1) maximize return, while (2) executing diverse skills. Compared with other Quality-Diversity methods, QDAC achieves significantly higher performance and more diverse behaviors on six challenging continuous control locomotion tasks. We also demonstrate that we can harness the learned skills to adapt better than other baselines to five perturbed environments. Finally, qualitative analyses showcase a range of remarkable behaviors: adaptive-intelligent-robotics.github.io/QDAC.} }
Endnote
%0 Conference Paper %T Quality-Diversity Actor-Critic: Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics %A Luca Grillotti %A Maxence Faldor %A Borja G. León %A Antoine Cully %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-grillotti24a %I PMLR %P 16416--16459 %U https://proceedings.mlr.press/v235/grillotti24a.html %V 235 %X A key aspect of intelligence is the ability to demonstrate a broad spectrum of behaviors for adapting to unexpected situations. Over the past decade, advancements in deep reinforcement learning have led to groundbreaking achievements to solve complex continuous control tasks. However, most approaches return only one solution specialized for a specific problem. We introduce Quality-Diversity Actor-Critic (QDAC), an off-policy actor-critic deep reinforcement learning algorithm that leverages a value function critic and a successor features critic to learn high-performing and diverse behaviors. In this framework, the actor optimizes an objective that seamlessly unifies both critics using constrained optimization to (1) maximize return, while (2) executing diverse skills. Compared with other Quality-Diversity methods, QDAC achieves significantly higher performance and more diverse behaviors on six challenging continuous control locomotion tasks. We also demonstrate that we can harness the learned skills to adapt better than other baselines to five perturbed environments. Finally, qualitative analyses showcase a range of remarkable behaviors: adaptive-intelligent-robotics.github.io/QDAC.
APA
Grillotti, L., Faldor, M., G. León, B. & Cully, A.. (2024). Quality-Diversity Actor-Critic: Learning High-Performing and Diverse Behaviors via Value and Successor Features Critics. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:16416-16459 Available from https://proceedings.mlr.press/v235/grillotti24a.html.

Related Material