Deep Deterministic Policy Gradient (DDPG)

Off-Policy Learning

Learning method where the agent learns an optimal policy while following another behavior policy, allowing for better exploration.

← Quay lại