(40 points) Part 1: Q- Learning and Policy Iteration on the Frozen Lake Environment In this part of the assignment, you will implement a basic version of Q-Learning on the Frozen Lake environment, following the tutorial provided here. The main purpose is to familiarize yourself with using the RL Gym library. In part 2, we will be using an environment of your choosing. Objective: Implement a Q-Learning agent to solve the Frozen Lake environment. Tasks: 1. Familiarize yourself with the Frozen Lake environment and its dynamics. 2. Implement the Q-Learning algorithm using the tutorial as a guide. 3. Train your Q-Learning agent on the Frozen Lake environment. 4. Evaluate the performance of your agent and analyze the impact of hyperparameters on the learning process. Specifically, verify the impact of the following hyperparameters: ⚫ alpha: learning rate o gamma: discount factor о epsilon: exploration rate Test at 。 epsilon exploration rate Test at least 3 different values for each hyperparameter and explain the effect of each hyperparameter on the learning process. 5. Implement the policy iteration algorithm and compare its performance to Q-Learning.

Solved: (40 points) Part 1: Q- Learning and Policy Iteration on the Frozen Lak

Home
questions and answers
40 points part 1 q learning and policy iteration on the frozen lake en

Question

(40 points) Part 1: Q- Learning and Policy Iteration on the Frozen Lake Environment In this part of the assignment, you will implement a basic version of Q-Learning on the Frozen Lake environment, following the tutorial provided here. The main purpose is to familiarize yourself with using the RL Gym library. In part 2,