Reinforcement learning and classical planning

Some algorithms in reinforcement learning and classical planing are performed in this version of the Pacman Game: a multi-player capture-the-flag variant of Pacman, where agents control both Pacman and ghosts in coordinated team-based strategies. The represents a team that will try to eat the food on the far side of the map, while defending the food on its home side. This is part of the Berkeley University contest: Pacman Capture the Flag.

Check the video here: