Reinforcement Learning and Q-Learning in Maze Runner Game
This project will implement the policy iteration and value iterations algorithms in the Maze Runner game.
The input of the program is a text file as represented below: The text file's first line indicates the number of rows and columns, respectively. Then there are rows*coloumns lines representing each cell of the game; every line consists of 4 numbers showing which sides of the cell are accessible or blocked in this pattern: North East South West. For example, 1 0 1 0 means we can only access the north and south sides. Then at the last line of the file, two numbers represent the agent's position.