| Project name | Topics | Description |
|---|---|---|
| search | Search (BFS, DFS, UCS, A*) | Find shortest path for pacman to eat all food |
| multiagent | Minimax | Maximize food being eaten while chased by ghosts |
| reinforcement | Q Learning | Use reinforcement learning to maximize food being eaten while chased by ghosts |
Run autograder
python autograder.py
Run pacman with A* agent, in bigMaze, using manhattan distance as heuristic
python pacman.py -l bigMaze -z .5 -p SearchAgent -a fn=astar,heuristic=manhattanHeuristic
Run autograder
python autograder.py
Run pacman with minimax agent, in smallClassic maze
python pacman.py -p AlphaBetaAgent -l smallClassic
Run autograder
python autograder.py
Run pacman
game 10 times with Q Learning agent, in smallGrid maze, after 2000 trainings
python pacman.py -p PacmanQAgent -x 2000 -n 2010 -l smallGrid
Pacman project structure and autograder was provided by UC Berkeley