Policies

• Categorical: Categorical policy for discrete action space.

• Categorical-MARL: Categorical policy for multi-agent system with discrete action space.

• Deterministic: Deterministic policy for continuous action space.

• Deterministic-MARL: Deterministic policy for multi-agent system with continuous action space.

• Gaussian: Gaussian policy for either discrtete or continous action spaces.

• Gaussian-MARL: Gaussian policy for multi-agent system with either discrete or continous action spaces.

• Mixiers: Mixers for value-decomposition based multi-agent reinforcement learning algorithms.

• Coordination-Graph: Tools of coordination graph for multi-agent reinforcement learning.