EasyDeepLearn

On-policy vs off-policy learning — what's the difference?

medium

Answer

  • On-policy methods learn from data generated by the current policy (e.g., SARSA, REINFORCE, PPO).
  • Off-policy methods learn from data collected by a different (behavior) policy — this enables replay buffers and reusing old data (e.g., Q-learning, DQN, SAC, DDPG).
  • Off-policy is more sample-efficient; on-policy is often more stable and easier to tune.
Check yourself — multiple choice
  • PPO is an off-policy algorithm
  • DQN is on-policy
  • Q-learning is off-policy; SARSA is on-policy
  • On-policy allows replay buffers

Q-learning bootstraps with maxa\operatorname{max}_{a}' Q(s',a') — off-policy. SARSA uses the actual next action — on-policy.

#fundamentals

Practise Reinforcement Learning

214 interview questions in this topic.

Related questions