Reinforcement Learning for Optimization

Actor-Critic Algorithm

Architecture combining an actor that selects actions according to a policy and a critic that evaluates these actions, enabling more stable and efficient learning.

← Wstecz