A policy in which the marginal costs of undertaking the policy equal the marginal benefits of that policy is best called an optimal policy.
What is an optimal policy?
- A policy that is as good as or better than all other policies is said to be an optimal policy.
- In other words, an ideal policy will be valued as highly as feasible in every state.
- There is usually at least one best course of action, but there could be more.
- The best course of action in a finite Markov Decision Process (MDP) is one that simultaneously maximizes the value of all the states.
- In other words, if an ideal policy exists, then the policy that maximizes state s' value and state s' value itself are equivalent.
To learn more about optimal policy refer,
https://brainly.com/question/17136362?referrer=searchResults
#SPJ4
