Generalized Policy Iteration

This is KEY to understanding how reinforcement learning works.

Generalized Policy Iteration refers to the general idea of letting policy-evaluation and policy-improvement processes interact.

Steps

  • Initialize policy
  • Repeat
    • Policy evaluation: Compute for model-free, for model-based
    • Policy improvement: update

Key Terminologies

Prediction / Learning policy evaluation Control policy-improvement

Some Ideas that I need to master

In David Silver’s diagrams, the up direction is to find , while the down arrows are to find .

With Model-Based problems, we use our full knowledge of the Markov Decision Process, using DP combined with Value Iteration or Value Iteration or Policy Iteration..? For the policy improvement, we use a policy that acts greedily with respect to V(s).

However, in Model-Free, where we don’t know the environment, we don’t have that luxury. Greedy policy improvement over V(s) requires model of MDP.

In model-free, we use . We reduce the burden. We can cache?

David Silver teaches two main methods: Monte-Carlo (full backup) and TD (bootstrap).