---

Week 9: Journal Questions

1. What is the main difference between reinforcement learning and supervised learning, and what is the appeal of using reinforcement learning?
2. Mitchell mentions a number of issues that reinforcement-learning researchers face. Pick one of these issues and explain what makes it problematic.
3. What is deep Q-learning? How does it differ from standard Q-learning?
4. Mitchell is skeptical that AI's focus on building game playing agents will lead to general-purpose and robust artificial intelligence, why?