20.500.12556/RUL-85003 Drevesno preiskovanje Monte Carlo s Thompsonovim vzorčenjem pri igri Prebivalci otoka Catan Monte Carlo Tree Search with Thompson sampling in The Settlers of Catan Drevesno preiskovanje Monte Carlo (MCTS) je ena izmed najbolj uporabljenih metod pri implementaciji močnega računalniškega igralca iger v umetni inteligenci, brez uporabe predhodnega znanja o domeni. Najmočnejši in najbolj popularni algoritmi, ki se pogosto uporabljajo za rešitev t.i. dileme raziskovanja (engl. exploration) proti izkoriščanju znanja (engl. exploitation) pri problemu več-rokih banditov, so raziskani in predstavljeni s pomočjo pregleda literature. Na podlagi empiričnih študij Thompsonovega vzorčenja v primerjavi s pristopom zgornje meje zaupanja (UCB) ter različicami podobnih algoritmov smo v magistrskem delu spremenili drevesno strategijo širjenja v MCTS. Končna domena aplikacije spremenjenega algoritma je družabna igra Prebivalci otoka Catan (SoC), implementirana v programskem jeziku C, skupaj z MCTS-UCT agentom, MCTS-TS agentom ter dvema preprosto igrajočima agentoma. Meritve učinkovitosti naštetih agentov prikazujejo povečano moč igranja agenta s spremenjeno drevesno strategijo, v primerjavi z najbolj pogosto uporabljenim pristopom, t.j. UCT. Monte Carlo Tree search (MCTS) is a popular method of choice for addressing the problem of a strong computer based game playing agent in Artificial Intelligence, without any prior domain knowledge. The strongest and most popular algorithms used to tackle the so-called exploration vs. exploitation dilemma in Multi-armed Bandit (MAB) problems were identified and presented in a literature review. Empirical studies measuring the performance of Thompson sampling (TS) and the state-of-the-art Upper Confidence Bound (UCB) approach in the classical MAB problem have been found, results of which support our modified tree policy in MCTS. The domain of application is the board game of the Settlers of Catan (SoC), which is implemented as a multi-agent environment in the programming language C, along with a MCTS-UCT agent, MCTS-TS agent and two strategy playing agents, namely the ore-grain and wood-clay agent. Performance measurements of the aforementioned agents, presented and discussed in this work, demonstrate an increase in the performance of the agent with the modified tree policy, when compared to the state-of-the-art approach (UCT). drevesno preiskovanje Monte Carlo (MCTS) več-roki ban- dit (MAB) zgornja meja zaupanja pri drevesih (UCT) Thompsonovo vzorčenje (TS) umetna inteligenca (AI) Prebivalci otoka Catan (SoC). Monte Carlo Tree Search (MCTS) Multi-armed Bandits (MAB) Upper Confidence Bound for Trees (UCT) Thompson sam- pling (TS) Artificial Intelligence (AI) the Settlers of Catan (SoC). true false false Slovenski jezik Angleški jezik Magistrsko delo/naloga 2016-09-09 09:45:01 2016-09-09 09:45:05 2022-08-02 06:44:39 0000-00-00 00:00:00 2016 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 17164 Tuma_Katja_-_Drevesno_preiskovanje_Monte_Carlo_s_Thompsonovim_vzorcenjem_pri_igri_Prebivalci_oto.pdf Tuma_Katja_-_Drevesno_preiskovanje_Monte_Carlo_s_Thompsonovim_vzorcenjem_pri_igri_Prebivalci_oto.pdf 1 EF6EDF14107F39080D67D70724B48C3F 12ca4b26eb7b35b033bbe12ad8458a59465424b8f207b13bbce045f748b9cb44 f1b85d1f-a1b0-11eb-a523-00155dcfd717 20.500.12556/rul/99812f5d-18bd-4d8a-85c2-5cc30581ed57 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=88121 Fakulteta za računalništvo in informatiko 0 0 0