General game playing with deep reinforcement learning
LE3 .A278 2021
2021
Lee, Greg
Acadia University
Bachelor of Science
Honours
Computer Science
Reinforcement learning provides problem formulations from the world of robotics and to arguably life itself. It also provides a set of solutions to such problems which may be explored in a computationally realistic manner with the help of neural networks, and thus we have deep reinforcement learning. As amazing as fields like robotics are to behold, the field of video games offers a more practical way of investigating the various deep reinforcement techniques. This thesis brings the concepts of deep reinforcement learning to Atari and NES games while keeping the approach as general as possible.
The author retains copyright in this thesis. Any substantial copying or any other actions that exceed fair dealing or other exceptions in the Copyright Act require the permission of the author.
https://scholar.acadiau.ca/islandora/object/theses:3567
