We propose a multiple-komi modification of the AlphaGo Zero/Leela Zero paradigm. The winrate as a function of the komi is modeled with a two-parameters sigmoid function, hence the winrate for all komi values is obtained, at the price of predicting just one more variable. A second novel feature is that training is based on self-play games that occasionaly branch -with changed komi- when the position is uneven. With this setting, reinforcement learning is shown to work on 7×7 Go, obtaining very strong playing agents. As a useful byproduct, the sigmoid parameters given by the network allow to estimate the score difference on the board, and to evaluate how much the game is decided. Finally, we introduce a family of agents which target winning moves with a higher score difference.

SAI a Sensible Artificial Intelligence that plays Go / Morandin, F.; Amato, G.; Gini, R.; Metta, C.; Parton, M.; Pascutto, G. -C.. - STAMPA. - 2019-(2019), pp. 1-8. ((Intervento presentato al convegno 2019 International Joint Conference on Neural Networks, IJCNN 2019 tenutosi a Budapest, Hungary nel 2019 [10.1109/IJCNN.2019.8852266].

SAI a Sensible Artificial Intelligence that plays Go

Morandin F.
Writing – Original Draft Preparation
;
2019

Abstract

We propose a multiple-komi modification of the AlphaGo Zero/Leela Zero paradigm. The winrate as a function of the komi is modeled with a two-parameters sigmoid function, hence the winrate for all komi values is obtained, at the price of predicting just one more variable. A second novel feature is that training is based on self-play games that occasionaly branch -with changed komi- when the position is uneven. With this setting, reinforcement learning is shown to work on 7×7 Go, obtaining very strong playing agents. As a useful byproduct, the sigmoid parameters given by the network allow to estimate the score difference on the board, and to evaluate how much the game is decided. Finally, we introduce a family of agents which target winning moves with a higher score difference.
SAI a Sensible Artificial Intelligence that plays Go / Morandin, F.; Amato, G.; Gini, R.; Metta, C.; Parton, M.; Pascutto, G. -C.. - STAMPA. - 2019-(2019), pp. 1-8. ((Intervento presentato al convegno 2019 International Joint Conference on Neural Networks, IJCNN 2019 tenutosi a Budapest, Hungary nel 2019 [10.1109/IJCNN.2019.8852266].
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11381/2870628
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? 1
social impact