MATLAB implementation of a few common Reinforcement Learning algorithms (value iteration, policy iteration, Q-learning, Sarsa, expected Sarsa) on a gridworld exploration task. This was done as an assignment in the context of the University of Cambridge's MPhil in Machine Learning and Machine Intelligence ( https://www.mlmi.eng.cam.ac.uk/ ). The code for the initialization of the gridworld (the files "cliffworld.m", "gridworld.m", "smallworld.m" and "initGridworld.m") was provided by the teaching staff. Additional information can be found in "RL_report.pdf".
Examples of experiments can be found in the files starting with "experiment_".
