Skaltsis, George MariosShin, HyosangTsourdos, Antonios2022-02-162022-02-162021-12-29Skaltsis GM, Shin H-S, Tsourdos A. (2021) Self play with parameter sharing in n-player mixed competitive-cooperative games. In: AIAA SciTech 2022 Forum, 3-7 January 2022, San Diego, CA and Virtual Eventhttps://doi.org/10.2514/6.2022-2498https://dspace.lib.cranfield.ac.uk/handle/1826/17576We introduce some parameter sharing multi-agent reinforcement learning schemes, combined with self-play for n-players mixed competitive-cooperative games. Except for the pure self-play scheme, an another one using the best policy, outperform the pure parameter sharing baseline, leading to better exploration of the state space and protecting from the performance deterioration observed with the baseline.enAttribution-NonCommercial 4.0 Internationalhttp://creativecommons.org/licenses/by-nc/4.0/Reinforcement LearningMarkov Decision ProcessStatistical DistributionsOptimization AlgorithmMulti Agent SystemComputer SystemsValue FunctionNeural NetworksMathematical ModelsGreedy AlgorithmConference paper