Which of the following are correct? Select all that apply.

In the tabular case, both policy iteration and value iteration are guaranteed to find an optimal policy.

Policy iteration always takes more policy improvement steps to reach convergence than value iteration

Value iteration can be done $($i$.$e made to converge to the correct optimal values$)$ asynchronously by updating a single state as long as all states are updated infinite times.

If value iteration has no error $($it converges to an exact value$)$ for a fixed tabular MDP$,$ it then always results in one fixed deterministic policy.

unanswered