chrisyuancn / cse511a-2017fall

Class assignments for Washington University's CSE 511A: Introduction to Artificial Intelligence

CSE 511A: Introduction to Artificial Intelligence – Fall 2017

This is the collection of class assignments for CSE 511A: Introduction to Artificial Intelligence – Fall 2017 in Washington University. It is based on CS188 at UC Berkeley.

Project 1: General search algorithms

Goals

Help Pacman agent to find paths through his maze world, both to reach a particular location and to collect food efficiently.

1. Depth-first search (DFS)
2. Breadth-first search (BFS)
3. Uniform-cost graph search
4. A* graph search
5. Implement a search problem CornersProblem
6. Implement a consistent, non-trivial heuristic for CornersProblem
7. Implement a consistent, non-trivial heuristic for FoodSearchProblem
8. Suboptimal search

Project 2: Adversarial search

Goals

Design agents for the classic version of Pacman, including ghosts.

1. Reflex agent
2. Minimax agent
3. Alpha-beta pruning
4. Expectimax agent
5. Better evaluation function

Project 3: Reinforcement learning

Goals

Implement value iteration and Q-learning.

1. Value iteration
2. Value iteration: parameter tuning #1
3. Value iteration: parameter tuning #2
4. Q-learning
5. Q-learning: epsilon-greedy
6. Q-learning: parameter tuning
7. Q-learning: crawler
8. Q-learning: training
9. Approximate Q-learning

Project 4: Probabilistic inference

Goals

Design Pac-Man agents that use sensors to locate and eat invisible ghosts.

1. Exact inference: stationary
2. Exact inference: moving
3. Greedy action
4. Particle filtering
5. Joint particle filtering: Bayesian network #1
6. Joint particle filtering: Bayesian network #2