MO-Gym: A Library of Multi-Objective Reinforcement Learning Environments
Published in BNAIC/BeNeLearn 22, 2022
Recommended citation: L. N. Alegre, F. Felten, E.-G. Talbi, G. Danoy, A. Nowé, and A. L. C. Bazzan, “MO-Gym: A Library of Multi-Objective Reinforcement Learning Environments,” Proceedings of the 34th Benelux Conference on Artificial Intelligence BNAIC/Benelearn 2022 https://bnaic2022.uantwerpen.be/wp-content/uploads/BNAICBeNeLearn_2022_submission_6485.pdf
We introduce MO-Gym, an extensible library containing a diverse set of multi-objective reinforcement learning environments. It introduces a standardized API that facilitates conducting experiments and performance analyses of algorithms designed to interact with multiobjective Markov decision processes. Importantly, it extends the widely used OpenAI Gym API, allowing the reuse of algorithms and features that are well-established in the reinforcement learning community. MO-Gym is available at: https://github.com/LucasAlegre/mo-gym.