60.1K

Verified Solution

Link Copied!

SELECT ALL THAT ARE TRUE

Consider the following Markov Decision Process (MDP): MDP with 4 states (rewards for each action are indicated on the arrow) There are 4 states A, B, C, and D. We can move up or down from states B and C, but only up for A and only down for D. Note that the discount factory = 0.75, and that this MDP is deterministic i.e. if you choose action UP, you are guaranteed to move UP, and likewise for action DOWN. Select all that are true In an MDP, the optimal policy for a given state s is unique The value iteration algorithm is solved recursively For a given MDP, the value function V* (s) of each state is known a priori V* (s) = [_T(s, a, s') [R(s, a, s') +yV* (s')] Q* (s,a) = [T(s,a, s') [R(s,a, s') + yV* (s')] Consider the following Markov Decision Process (MDP): MDP with 4 states (rewards for each action are indicated on the arrow) There are 4 states A, B, C, and D. We can move up or down from states B and C, but only up for A and only down for D. Note that the discount factory = 0.75, and that this MDP is deterministic i.e. if you choose action UP, you are guaranteed to move UP, and likewise for action DOWN. Select all that are true In an MDP, the optimal policy for a given state s is unique The value iteration algorithm is solved recursively For a given MDP, the value function V* (s) of each state is known a priori V* (s) = [_T(s, a, s') [R(s, a, s') +yV* (s')] Q* (s,a) = [T(s,a, s') [R(s,a, s') + yV* (s')]

Answer & Explanation Solved by verified expert

See Answer