Reinforcement Learning for Optimization

Policy Optimization

Class of methods in reinforcement learning that directly optimize the policy without going through a value function, often using policy gradient techniques.

← Zurück