Model-based RL in ATARI games
This report describes the implementation of the world model reinforcement learning algorithm to gauge its performance against traditional reinforcement learning algorithms like deep Q-learning. The algorithm will be tested in different Atari Environments, some of which pose great difficultly due to...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/148080 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This report describes the implementation of the world model reinforcement learning algorithm to gauge its performance against traditional reinforcement learning algorithms like deep Q-learning. The algorithm will be tested in different Atari Environments, some of which pose great difficultly due to the smaller signals in the observation frames. Specific modifications were required to tailor the world model for each of the different environments. |
|---|