Mini-project suggestions

What successful applications are there for reinforcement learning? What techniques did they use?

What techniques exist that mix model-based reinforcement learning and features? Explain to the class how they work.