Using Reinforcement Learning for ball balancing task (Hysteretic Q-Learning)

Hi everyone,
Has anyone implemented a multi-agent reinforcement algorithm like the one in this paper?
an_algorithm_for_Decentralized_Reinforcement_Learning_in_Cooperative.pdf (510.5 KB)
Check out the “Experiments on a collaborative ball balancing task” part. I want to know how I can understand and implement this kind of task, especially how to discretize continuous action spaces such as in this scenario.

1 Like