Single-player to Two-player Knowledge Transfer in Atari 2600 Games

Saadat, Kimiya

Single-player to Two-player Knowledge Transfer in Atari 2600 Games

dc.contributor.advisor	Zhao, Richard
dc.contributor.author	Saadat, Kimiya
dc.contributor.committeemember	Abou-Zeid, Hatem
dc.contributor.committeemember	Aycock, John
dc.date	2025-02-13
dc.date.accessioned	2024-11-21T16:38:26Z
dc.date.available	2024-11-21T16:38:26Z
dc.date.issued	2024-11-18
dc.description.abstract	Playing two-player games using reinforcement learning and self-play can be challenging due to the complexity of two-player environments and the potential instability in the training process. It is proposed that a reinforcement learning algorithm can train more efficiently and achieve improved performance in a two-player game by leveraging the knowledge from the single-player version of the same game. This study examines the proposed idea in ten different Atari 2600 environments using the Atari 2600 RAM as the input state. The advantages of using transfer learning from a single-player training process over training in a two-player setting from scratch are discussed, and the results are demonstrated in several metrics, such as the training time and average total reward. Finally, a method for calculating RAM complexity and its relationship to performance after transfer is discussed. Results show that in most cases transferred agent is performing better than the agent trained from scratch while taking less time to train. Moreover, it is shown that RAM complexity can be used as a weak predictor to predict the transfer's effectiveness.
dc.identifier.citation	Saadat, K. (2024). Single-player to two-player knowledge transfer in Atari 2600 games (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.
dc.identifier.uri	https://hdl.handle.net/1880/120090
dc.language.iso	en
dc.publisher.faculty	Graduate Studies
dc.publisher.institution	University of Calgary
dc.rights	University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.
dc.subject	Reinforcement Learning
dc.subject	Transfer Learning
dc.subject	Game-Playing AI
dc.subject	Atari 2600
dc.subject	Deep Q Networks
dc.subject	Multiplayer Games
dc.subject	Deep Reinforcement Learning
dc.subject	Deep Learning
dc.subject.classification	Computer Science
dc.subject.classification	Artificial Intelligence
dc.title	Single-player to Two-player Knowledge Transfer in Atari 2600 Games
dc.type	master thesis
thesis.degree.discipline	Computer Science
thesis.degree.grantor	University of Calgary
thesis.degree.name	Master of Science (MSc)
ucalgary.thesis.accesssetbystudent	I do not require a thesis withhold – my thesis will have open access and can be viewed and downloaded publicly as soon as possible.