<< mdp_eval_policy_iterative Markov Decision Processses (MDP) Toolbox mdp_eval_policy_optimali >>

Markov Decision Processses (MDP) Toolbox >> Markov Decision Processses (MDP) Toolbox > mdp_eval_policy_matrix

mdp_eval_policy_matrix

Evaluates a policy using matrix operation.

Calling Sequence

Vpolicy = mdp_eval_policy_matrix (P, R, discount, policy)

Description

mdp_eval_policy_matrix evaluates the value fonction associated with a policy.

Arguments

P

transition probability array.

P can be a 3 dimensions array (SxSxA) or a list (1xA), each list element containing a sparse matrix (SxS).

R

reward array.

R can be a 3 dimensions array (SxSxA) or a list (1xA), each list element containing a sparse matrix (SxS) or a 2D array (SxA) possibly sparse.

discount

discount factor

discount is a real which belongs to [0; 1[.

policy

a policy.

policy is a (Sx1) vector. Each element is an integer corresponding to an action.

Evaluation

Vpolicy

value fonction.

Vpolicy is a (Sx1) vector.

Examples

-> P = list();
-> P(1) = [ 0.5 0.5;   0.8 0.2 ];
-> P(2) = [ 0 1;   0.1 0.9 ];
-> R = [ 5 10;   -1 2 ];

-> Vpolicy = mdp_eval_policy_matrix(P, R, 0.9, [1; 2])
Vpolicy =
   28.90625
   24.21875

In the above example, P can be a list containing sparse matrices:
-> P(1) = sparse([ 0.5 0.5;  0.8 0.2 ]);
-> P(2) = sparse([ 0 1;  0.1 0.9 ]);
The function is unchanged.

Authors


Report an issue
<< mdp_eval_policy_iterative Markov Decision Processses (MDP) Toolbox mdp_eval_policy_optimali >>