{"id":{"repo_id":"vilnius","oai_identifier":"oai:vu.lt:elaba:308337049"},"canonical_url":"https://search.ndltd.org/etd/vilnius/oai:vu.lt:elaba:308337049","repository":{"repo_id":"vilnius","name":"Vilnius University","base_url":"https://epublications.vu.lt/oai"},"display":{"title":"Monte Karlo medžių paieškos algoritmų ir neuroninių tinklų pritaikymas žaidimo agentuose /","abstract":"This work explores the creation of game agents for board game \"Azul\" using Monte Carlo tree search and neural networks. In this work the process of game agents' creation, the principles of their operation and design details are described. During this work various strategies of integrating neural networks into Monte Carlo tree search were tested, comparisons between different neural network architectures were executed. Also, created agents were tested by playing them against each other and compared against agents that use different algorithms. It is shown how integrating neural networks with Monte Carlo tree search algorithm can make it possible to achieve better playing strength than by using them separately. Moreover, these agents can reliably beat heuristic agent and agent, based on classical Monte Carlo tree seach algorithm. Furthermore, after comparing different neural network architectures it was found that smaller neural networks have an advantage over bigger ones in improving the performance of game agents that use Monte Carlo tree search. It is also shown that an accurate evaluation of a position is very important to the quality of Monte Carlo tree search based agent's game playing ability.","abstract_html":"This work explores the creation of game agents for board game &quot;Azul&quot; using Monte Carlo tree search and neural networks. In this work the process of game agents&#x27; creation, the principles of their operation and design details are described. During this work various strategies of integrating neural networks into Monte Carlo tree search were tested, comparisons between different neural network architectures were executed. Also, created agents were tested by playing them against each other and compared against agents that use different algorithms. It is shown how integrating neural networks with Monte Carlo tree search algorithm can make it possible to achieve better playing strength than by using them separately. Moreover, these agents can reliably beat heuristic agent and agent, based on classical Monte Carlo tree seach algorithm. Furthermore, after comparing different neural network architectures it was found that smaller neural networks have an advantage over bigger ones in improving the performance of game agents that use Monte Carlo tree search. It is also shown that an accurate evaluation of a position is very important to the quality of Monte Carlo tree search based agent&#x27;s game playing ability.","abstract_has_math":false,"creators":["Jarockis, Augustinas,"],"institution":"Institutional Repository of Vilnius University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-19T10:12:59Z","subjects":[],"languages":["lit"],"rights":["info:eu-repo/semantics/openAccess"],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://repository.vu.lt/VU:ELABAETD308337049&prefLang=en_US","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://repository.vu.lt/VU:ELABAETD308337049&prefLang=en_US","record_label":"Repository record","record_kind":"record","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":"open","embargo_until":null,"access_note":null},"source_record":{"url":"https://epublications.vu.lt/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Avu.lt%3Aelaba%3A308337049","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Jarockis, Augustinas,"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:publisher","label":"Institution","values":["Institutional Repository of Vilnius University"]},{"key":"dc:relation","label":"Dc Relation","values":["https://epublications.vu.lt/object/elaba:308337049/308337049.pdf"]},{"key":"dc:type","label":"Dc Type","values":["info:eu-repo/semantics/bachelorThesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["lit"]},{"key":"dc:rights","label":"Dc Rights","values":["info:eu-repo/semantics/openAccess"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://repository.vu.lt/VU:ELABAETD308337049&prefLang=en_US"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["This work explores the creation of game agents for board game \"Azul\" using Monte Carlo tree search and neural networks. In this work the process of game agents' creation, the principles of their operation and design details are described. During this work various strategies of integrating neural networks into Monte Carlo tree search were tested, comparisons between different neural network architectures were executed. Also, created agents were tested by playing them against each other and compared against agents that use different algorithms. It is shown how integrating neural networks with Monte Carlo tree search algorithm can make it possible to achieve better playing strength than by using them separately. Moreover, these agents can reliably beat heuristic agent and agent, based on classical Monte Carlo tree seach algorithm. Furthermore, after comparing different neural network architectures it was found that smaller neural networks have an advantage over bigger ones in improving the performance of game agents that use Monte Carlo tree search. It is also shown that an accurate evaluation of a position is very important to the quality of Monte Carlo tree search based agent's game playing ability."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Monte Karlo medžių paieškos algoritmų ir neuroninių tinklų pritaikymas žaidimo agentuose /","The application of monte carlo tree search algorithms and neural networks for game agents."]}]}],"canonical_facts":{"dc:creator":["Jarockis, Augustinas,"],"dc:date":["2026"],"dc:description":["This work explores the creation of game agents for board game \"Azul\" using Monte Carlo tree search and neural networks. In this work the process of game agents' creation, the principles of their operation and design details are described. During this work various strategies of integrating neural networks into Monte Carlo tree search were tested, comparisons between different neural network architectures were executed. Also, created agents were tested by playing them against each other and compared against agents that use different algorithms. It is shown how integrating neural networks with Monte Carlo tree search algorithm can make it possible to achieve better playing strength than by using them separately. Moreover, these agents can reliably beat heuristic agent and agent, based on classical Monte Carlo tree seach algorithm. Furthermore, after comparing different neural network architectures it was found that smaller neural networks have an advantage over bigger ones in improving the performance of game agents that use Monte Carlo tree search. It is also shown that an accurate evaluation of a position is very important to the quality of Monte Carlo tree search based agent's game playing ability."],"dc:format":["application/pdf"],"dc:identifier":["https://repository.vu.lt/VU:ELABAETD308337049&prefLang=en_US"],"dc:language":["lit"],"dc:publisher":["Institutional Repository of Vilnius University"],"dc:relation":["https://epublications.vu.lt/object/elaba:308337049/308337049.pdf"],"dc:rights":["info:eu-repo/semantics/openAccess"],"dc:title":["Monte Karlo medžių paieškos algoritmų ir neuroninių tinklų pritaikymas žaidimo agentuose /","The application of monte carlo tree search algorithms and neural networks for game agents."],"dc:type":["info:eu-repo/semantics/bachelorThesis"]},"updated_at":"2026-11-19T10:12:59Z"}