Many environments in which an agent can use reinforcement learning techniques to learn profitable strategies are affected by other learning agents. These situations can be modeled as general-sum games. When playing repeated general-sum games with other learning agents, the goal of a self-interested learning agent is to maximize its own payoffs over time. Traditional reinforcement learning algorithms learn myopic strategies in these games. As a result, they learn strategies that produce undesirable results in many games. In this dissertation, we develop and analyze algorithms that learn non-myopic strategies when playing many important infinitely repeated general-sum games. We show that, in many of these games, these algorithms outperform existing multiagent learning algorithms. We derive performance guarantees for these algorithms (for certain learning parameters) and show that these guarantees become stronger and apply to larger classes of games as more information is observed and used by the agents. We establish these results through empirical studies and mathematical proofs.
","abstract_html":"<p>Many environments in which an agent can use reinforcement learning techniques to learn profitable strategies are affected by other learning agents. These situations can be modeled as general-sum games. When playing repeated general-sum games with other learning agents, the goal of a self-interested learning agent is to maximize its own payoffs over time. Traditional reinforcement learning algorithms learn myopic strategies in these games. As a result, they learn strategies that produce undesirable results in many games. In this dissertation, we develop and analyze algorithms that learn non-myopic strategies when playing many important infinitely repeated general-sum games. We show that, in many of these games, these algorithms outperform existing multiagent learning algorithms. We derive performance guarantees for these algorithms (for certain learning parameters) and show that these guarantees become stronger and apply to larger classes of games as more information is observed and used by the agents. We establish these results through empirical studies and mathematical proofs.</p>","abstract_has_math":false,"creators":["Crandall, Jacob W."],"institution":"Brigham Young University - Provo","degree_name":"PhD","degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-07-24T01:27:43Z","subjects":["multi-agent learning","reinforcement learning","satisficing","Computer Sciences"],"languages":["English"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://scholarsarchive.byu.edu/etd/337","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Crandall, Jacob W."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2005-12-21T08:00:00Z"]},{"key":"dc:publisher","label":"Institution","values":["Brigham Young University - Provo"]},{"key":"dc:type","label":"Dc Type","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["PhD"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["multi-agent learning","reinforcement learning","satisficing","Computer Sciences"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["English"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarsarchive.byu.edu/etd/337","https://scholarsarchive.byu.edu/context/etd/article/1336/viewcontent/ETD_CISOPTR_557.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Physical and Mathematical Sciences; Computer Science"]},{"key":"dc:description.abstract","label":"Abstract","values":["Many environments in which an agent can use reinforcement learning techniques to learn profitable strategies are affected by other learning agents. These situations can be modeled as general-sum games. When playing repeated general-sum games with other learning agents, the goal of a self-interested learning agent is to maximize its own payoffs over time. Traditional reinforcement learning algorithms learn myopic strategies in these games. As a result, they learn strategies that produce undesirable results in many games. In this dissertation, we develop and analyze algorithms that learn non-myopic strategies when playing many important infinitely repeated general-sum games. We show that, in many of these games, these algorithms outperform existing multiagent learning algorithms. We derive performance guarantees for these algorithms (for certain learning parameters) and show that these guarantees become stronger and apply to larger classes of games as more information is observed and used by the agents. We establish these results through empirical studies and mathematical proofs.
"]},{"key":"dc:format","label":"Dc Format","values":["application:pdf"]},{"key":"dc:source","label":"Dc Source","values":["Brigham Young University - Provo"]},{"key":"dc:title","label":"Title","values":["Learning Successful Strategies in Repeated General-sum Games"]}]}],"canonical_facts":{"dc:creator":["Crandall, Jacob W."],"dc:date":["2005-12-21T08:00:00Z"],"dc:description":["Physical and Mathematical Sciences; Computer Science"],"dc:description.abstract":["Many environments in which an agent can use reinforcement learning techniques to learn profitable strategies are affected by other learning agents. These situations can be modeled as general-sum games. When playing repeated general-sum games with other learning agents, the goal of a self-interested learning agent is to maximize its own payoffs over time. Traditional reinforcement learning algorithms learn myopic strategies in these games. As a result, they learn strategies that produce undesirable results in many games. In this dissertation, we develop and analyze algorithms that learn non-myopic strategies when playing many important infinitely repeated general-sum games. We show that, in many of these games, these algorithms outperform existing multiagent learning algorithms. We derive performance guarantees for these algorithms (for certain learning parameters) and show that these guarantees become stronger and apply to larger classes of games as more information is observed and used by the agents. We establish these results through empirical studies and mathematical proofs.
"],"dc:format":["application:pdf"],"dc:identifier":["https://scholarsarchive.byu.edu/etd/337","https://scholarsarchive.byu.edu/context/etd/article/1336/viewcontent/ETD_CISOPTR_557.pdf"],"dc:language":["English"],"dc:publisher":["Brigham Young University - Provo"],"dc:source":["Brigham Young University - Provo"],"dc:subject":["multi-agent learning","reinforcement learning","satisficing","Computer Sciences"],"dc:title":["Learning Successful Strategies in Repeated General-sum Games"],"dc:type":["Dissertation"],"thesis:degree_name":["PhD"]},"updated_at":"2026-07-24T01:27:43Z"}