JC01111/Pacman-AI-Projects
Developed advanced algorithms (A*, Markov Decision Process, Reinforcement Learning, and Machine Learning) to optimize Pac-Man’s path in the maze for efficient pellet collection while evading ghosts.
Pac-Man AI Projects
Overview
These are the main topics of the Pac-Man AI Projects.
I implemented many advanced Searching, Reinforcement Learning, and Machine Learning algorithms for Pac-Man to plan out paths to efficiently eat all the dots while evading ghosts.
Please click on the topics you are interested and read the README.md file. I've labeled the files I implemented of algorithms, and the README.md file will demonstrate the differences and effects for different algorithms.
Topics
Search
Implement depth-first, breadth-first, uniform cost, and A* search algorithms. These algorithms are used to solve navigation and traveling salesman problems in the Pacman world.
Multi-Agent Search
Classic Pacman is modeled as both an adversarial and a stochastic search problem. Implement multiagent minimax and expectimax algorithms, as well as designing evaluation functions.
Reinforcement Learning
Implement Value Function, Q learning, and Approximate Q learning to help pacman and crawler agents learn rational policies.
Bayes Nets and Hidden Markov Model
Pacman uses probabilistic inference on Bayes Nets and the forward algorithm and particle sampling in a Hidden Markov Model to find ghosts given noisy readings of distances to them.
Machine Learning
Implement the perceptron algorithm, neural network, and recurrent nn models, and apply the models to several tasks including digit classification and language identification.