Artificial Intelligence   Total Points: 50   Write a program to define an MDP (Markov’s Decision Process) problem for a generic grid world (e

Artificial Intelligence

Total Points: 50

Write a program to define an MDP (Markov’s Decision Process) problem for a generic grid world (e.g. figure 17.1 of AIMA Textbook 4th edition page 563) and use value iteration (figure 17.6) to print the values of states in each iteration. After termination of the value iteration, the final policy should be printed. Next, implement a modified policy iteration (policy iteration algorithm is provided in figure 17.9 and modified policy iteration is discussed on page 578) to calculate and print the optimal policy. The description of MDP (T, R, gamma, and epsilon) should be loaded from the text file as input. You can use the sample input format included with the assignment or use your own. Also, write-up describing the format of the input file and how to run the program is not necessary.

Programming Language

Python

Grading Rubric

Parse input text into MDP class => 5

Value Iteration => 15

Print values of states in each iteration=> 5

Policy Iteration => 15

Print final policy => 5

Writeup => 5