cs181-homework6

Spring 2021 Machine Learning (CS 181) Homework 6

Problem Topics

Solutions contained in the personal-solutions folder

1. "Explaining Away" effect in Bayesian Network Inference
2. Policy Iteration and Value Iteration in solving a Markov Decision Process (MDP)
3. Reinforcement Learning and implementing a Q-learning agent

---

Code

Implementation contained in the code folder

problem2-Policy-Value-Iteration.py

• Referred to as T6_P2.py in the specifications
• Implements Policy Iteration and Value Iteration to find the optimal solution to a Markov Decision Process (MDP)

problem3-Reinforcement-QLearning.py

• Referred to as stub.py in the specifications
• Implement a Q-learning agent to play a game that is similar to Flappy Bird