peterjiawhite/Superhuman_AI_in_multiplayer_poker_-
该论文主要介绍了美国卡内基梅隆大学团队,在多人德州扑克上的人工智能新思路,即不再简单寻找纳什均衡,而引入悔恨值的概念,自我博弈,并采用蒙特卡洛CFR方法,构建蓝图,该方法通用性强,该团队声称他们的德州扑克蓝图只在两枚CPU运算8天即可得出蓝图,即可以实现实时博弈。现已经有国内团队将其用在了斗地主上面,成效显著。
Python
Stargazers
- DongdongBaiInstitute of Automation, Chinese Academy of Sciences,Beijing,China
- L-sure
- chandler-joye
- lpyvvvvvvChina
- lsxh
- aili1024
- chenkevin
- IamHuskarhenan province
- Michael-ZBeiJing
- xmzh1999
- HaHaEasonChina
- aqua4freedom
- blackjack911China
- voidFFF
- zixing131
- liuaqcsu
- sssirus
- flysosoEarth
- chpeagle7
- salyang
- 380799631
- 893403159
- SJ00425
- CHEN-CHI-TINTaiwan
- LouissMaHongKong