Final Project Ideas

• Write an interactive reinforcement-learning simulator.
• Apply techniques from statistical natural language processing to pair German and English news articles.
• Solve my cube puzzle (techniques from group theory or Korf's approach might help).
• Solve deterministic planning problems using informed search via linear programming.
• Apply machine learning or other AI techniques to solve systems problems.
• Use Bayes net algorithm to evaluate a probabilistic STRIPS plan.
• Devise a scheme for stochastic search in plan space. Compare to SATPLAN.
• Generate good probabilistic policies for unobservable domains like grid worlds (woods7).
• Find optimal memory-based policies for unobservable deterministic domains like grid world (woods7) using A*.
• Solve a huge MDP like racetrack using techniques from numerical analysis.
• Use hill-climbing, simulated annealing, and/or genetic algorithms to find good policies.
• Encode and solve Tetris. Or Pac Man.
• Create a problem that is hard for policy iteration.