Jay-Down/Berkeley-AI-Pac-Man-Agents
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|
Repository files navigation
AI agents for the version of the classic Pacman arcade game as developed at UC Berkeley (http://ai.berkeley.edu). 1. classifierAgents.py: builds a decision tree classifier, then implements an ensemble of these trees via bootstrap aggregation in order to return a single classification label as agreed upon by a maximum vote of the ensemble. 2. mlLearningAgents.py: An exploratory SARSA agent that actively learns Q values for each state-action pair in the Pacman smallGrid environment