五子棋吧 关注:53,097贴子:695,134
  • 16回复贴,共1

2022年五子棋锦标赛开始了(Gomocup)请大家集中在这个帖子里面

只看楼主收藏回复

Gomocup Schedule, 15th-17th of April 2022
There will be 2 freestyle groups (i.e., Freestyle 1 and 2), 1 fastgame group (Fastgame), 2 standard groups (i.e., Standard 1 and 2), and 1 renju group (Renju) in Gomocup 2022. Elo rating system is used to evaluate AIs' strength. The memory limit is determined to 350MB, and the AI size limitation is 8MB. Time per move/per match will be 5s/120s for fastgame, 300s/1000s for final leagues (i.e., Freestyle 1, Standard 1, and Renju), and 30s/180s for the rest (i.e., Freestyle 2 and Standard 2).
For Freestyle 2 and Standard 2, the top 4 AIs will move up to the next group (i.e., Freestyle 1 and Standard 1, respectively). If the top k (k>4) places were all taken by new (or updated) AIs in a group, then all these k AIs would advance to the next group.
In addition, there will be an unlimited tournament, where AIs will be running on participants' computers. This experimental tournament has no limitation on AI size and allows participants to use any hardware (e.g., GPUs) that they have access to. Time per match will be 120 min. There is no additional time limitation per move. Following Gomoku World Championships, the rule for this tournament is Swap2. The tournament will be a double round-robin tournament and tie-break criteria are: game points; match points; Berger coefficient; personal score, following World Correspondence Gomoku Championships.
Schedule (tentative)
Freestyle 2 - Saturday 12:30 AM UTC/GMT
Standard 2
Fastgame
Standard 1
Renju
Freestyle 1
What is new?Updates
AlphaGomoku (Kozarzewski) 2022 - Now using Monte Carlo Graph Search. Fixes and optimizations in the VCF solver. Improved time management. Added several utilities like benchmarking or automatic configuration.
Embryo 2022 - The evaluation also takes into account 3 + 2 and 2 + 2 patterns. C++ code ready to run on Android, release with GUI coming soon.
PentaZenNN 2022 - Switch to MCTS with NN evaluation modified from KataGo. The alpha-beta search serves as a parallel surewin finder now. PentaZenNN supports standard and renju rules.
Rapfi 2022 - Rapfi is now equipped with an NNUE (Efficiently updatable neural network) trained on games played by Katagomo. Although NNUE runs with a lower speed, its accuracy on evaluation brings notable strength improvement. Classical evaluation is improved via automatic tuning, and combines with NNUE in a hybrid way. Search and speed is also improved.
New AI
Angel - a Gomoku program developed by Yuan Xie. Negamax Alpha-Beta search, possible position sort, quick believable VCF search, Iterative Deepening. Angel supports freestyle rule.
Arbillo - a Gomoku program developed by Lorenzo Piazzo. First attempt. Standard (alpha-beta) player. Arbillo supports freestyle rule.
DDQK-Conquer - a Gomoku program developed by Sui Shunyi. Mainly using the use-value network algorithm, alpha beta screening is also involved. DDQK-Conquer supports freestyle rule.
Zhoumian - a Gomoku program developed by Shangren Shang. A low-IQ AI based on C language. Zhoumian supports freestyle rule.
LeaguesFreestyle 1
EMBRYO22
BARBAKAN
KATAGOMO21.E
RAPFI22
YIXIN18
ALPHAGOMOKU.MK22
PENTAZEN21.20
XOXO21.F
STARDUST
RENJUSOLVER.F
+ 4 best from Freestyle 2
Freestyle 2
WINE18
WHOSE19
SLOWRENJU20
HEWER18
TITO14
DEEPFIRE19
SWINE17
GOFIVE
CARBON17
ONIX16
EGO21
KANEC09
PELA17
GMOTOR11
CHIS17
ZETOR17
PECUCHET
EULRING
JUDE
SPARKLE
MONKEY
PUREGM21
QMENTAT6
DJALL18
NOESIS
PUSKVOREC
IMRO15
LICHT
FASTGOMOKU14
PISQ7
FIVEROW
STAHLFAUST
LIGHTBLUE
VALKYRIE
CRUSHER
PROLOG
AIBB
MUSHROOM
WHATATEAM
ANGEL
ARBILLO
DDQKCONQUER
ZHOUMIAN
Fastgame League
EMBRYO22
BARBAKAN
RAPFI22
KATAGOMO21.F
ALPHAGOMOKU.MK22
PENTAZEN21.20
YIXIN18
XOXO21.F
HEWER18
WHOSE19
SLOWRENJU20
TITO14
STARDUST
WINE18
SWINE17
DEEPFIRE19
CARBON17
GOFIVE
EGO21
KANEC09
GMOTOR11
PELA17
CHIS17
ZETOR17
PECUCHET
EULRING
LICHT
JUDE
ALPHAGOMOKU.GS
SPARKLE
MONKEY
PUREGM21
QMENTAT6
DJALL18
PUSKVOREC
NOESIS
IMRO15
FASTGOMOKU14
PISQ7
STAHLFAUST
FIVEROW
VALKYRIE
LIGHTBLUE
CRUSHER
PROLOG
AIBB
MUSHROOM
WHATATEAM
ANGEL
ARBILLO
DDQKCONQUER
ZHOUMIAN
Standard 1
KATAGOMO21.S
RAPFI22
EMBRYO22
ALPHAGOMOKU.MK22
PENTAZENNN22
YIXIN18
XOXO21.S
RENJUSOLVER.S
+ 4 best from Standard 2
Standard 2
TITO14
WHOSE19
HEWER18
SLOWRENJU20
DEEPFIRE19
GOFIVE
CARBON17
GMOTOR11
ZETOR17
PELA17
Renju League
KATAGOMO21.R
EMBRYO22
RAPFI22
YIXIN18
PENTAZENNN22
RENJUSOLVER.R
SLOWRENJU20
WHOSE19
GOFIVE
CARBON17
XL
PELA17
Unlimited Tournament
ALPHAGOMOKU.MK22
KATAGOMO21.E
PENTAZENNN22
RAPFI22


IP属地:陕西1楼2022-04-16 14:59回复
    看点:KATAGOMO21.E 参加所有的比赛。


    IP属地:陕西2楼2022-04-16 16:56
    回复
      看点2:Rapfi 2022 - Rapfi is now equipped with an NNUE。Rapfi加入了NNue评估,棋力大增,看看这次比赛的结果,估计能有不错的成绩。


      IP属地:陕西3楼2022-04-16 16:57
      回复
        胚胎看介绍并没有什么实质性的提升啊。
        Embryo 2022 - The evaluation also takes into account 3 + 2 and 2 + 2 patterns. C++ code ready to run on Android, release with GUI coming soon. 但是后续会出安卓版本,值得期待!


        IP属地:陕西4楼2022-04-16 16:58
        回复
          现在是无禁手和标准的乙组比赛,乙组前4名进入甲组比赛。甲组有看头!乙组有的引擎非常弱!


          IP属地:陕西5楼2022-04-16 17:02
          回复
            Updates
            AlphaGomoku (Kozarzewski) 2022 - Now using Monte Carlo Graph Search. Fixes and optimizations in the VCF solver. Improved time management. Added several utilities like benchmarking or automatic configuration.
            Embryo 2022 - The evaluation also takes into account 3 + 2 and 2 + 2 patterns. C++ code ready to run on Android, release with GUI coming soon.
            PentaZenNN 2022 - Switch to MCTS with NN evaluation modified from KataGo. The alpha-beta search serves as a parallel surewin finder now. PentaZenNN supports standard and renju rules.
            Rapfi 2022 - Rapfi is now equipped with an NNUE (Efficiently updatable neural network) trained on games played by Katagomo. Although NNUE runs with a lower speed, its accuracy on evaluation brings notable strength improvement. Classical evaluation is improved via automatic tuning, and combines with NNUE in a hybrid way. Search and speed is also improved.


            IP属地:陕西6楼2022-04-16 17:06
            回复
              New AI
              Angel - a Gomoku program developed by Yuan Xie. Negamax Alpha-Beta search, possible position sort, quick believable VCF search, Iterative Deepening. Angel supports freestyle rule.
              Arbillo - a Gomoku program developed by Lorenzo Piazzo. First attempt. Standard (alpha-beta) player. Arbillo supports freestyle rule.
              DDQK-Conquer - a Gomoku program developed by Sui Shunyi. Mainly using the use-value network algorithm, alpha beta screening is also involved. DDQK-Conquer supports freestyle rule.
              Zhoumian - a Gomoku program developed by Shangren Shang. A low-IQ AI based on C language. Zhoumian supports freestyle rule.


              IP属地:陕西7楼2022-04-16 17:06
              回复
                504 Gateway Time-outnginx/1.18.0 (Ubuntu)


                IP属地:陕西8楼2022-04-16 18:12
                回复
                  有没有详细赛制规则介绍光子赞助的机器没中文


                  IP属地:意大利来自Android客户端9楼2022-04-17 08:15
                  收起回复
                    我来推荐一个今年Gomocup中的一个新程序:DDQK-Conquer
                    虽然Fs2和fastgame都是倒数第十,elo评分1200
                    但是速度是真的快
                    出棋速度不超过0.001秒,可谓神速。
                    他的作者,在 https://user.gomocup.org/user/164/ 中写道:“A adorable boy from China, 14 years old.”
                    而且还提供了一个GUI界面
                    下载链接:https://github.com/duoduo666/DDQK-Conquer


                    IP属地:上海10楼2022-04-19 12:29
                    收起回复