Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 599 Bytes

README.md

File metadata and controls

9 lines (6 loc) · 599 Bytes

pacmanMachineLearning

Overview

Contains different code sections for training a pacman agent

Deep Q-networks are used to train pacman agent using higher order action relative inputs studied in this paper actions are then associated with rewards and using epsilon greedy algorithms to train the agent over a long period of time to create an efficient agent

Experimentation using NEAT-ai to create a better agent through the use of natural selection and random topologies of networks for the models instead of traditional reinforcement learning