Chess is one of the most studied fields in the history of artificial intelligence. Among the many methods with which this game is approached, the reinforcement learning strategy has already achieved important results, proving itself more effective than other engines that use different strategies. What makes this method special is the choice not to always take the best action but from time to time choose a path yet to be explored and from there try to obtain the results that lead to the greatest possible rewards.
Gli scacchi sono da sempre uno degli argomenti più studiati nella storia dell’intelligenza artificiale. Tra i vari metodi con cui questo gioco viene approcciato, la strategia del reinforcement learning ha già ottenuto risultati molto importanti, risultando più efficace di altri motori che utilizzano diverse strategie. Ciò che rende particolare questo metodo è la scelta di non compiere sempre l’azione migliore ma ogni tanto scegliere una strada ancora da esplorare e da lì provare ad ottenere i risultati che portano alle ricompense maggiori possibili.
Reinforcement Learning e le sue applicazioni: gli scacchi.
CALLEGARO, LUCA
2021/2022
Abstract
Chess is one of the most studied fields in the history of artificial intelligence. Among the many methods with which this game is approached, the reinforcement learning strategy has already achieved important results, proving itself more effective than other engines that use different strategies. What makes this method special is the choice not to always take the best action but from time to time choose a path yet to be explored and from there try to obtain the results that lead to the greatest possible rewards.File | Dimensione | Formato | |
---|---|---|---|
Callegaro_Luca.pdf
accesso aperto
Dimensione
1.37 MB
Formato
Adobe PDF
|
1.37 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/37991