DeepZenGo is the official name of newer versions of the Zen go program since deep learning techniques were added following the success of AlphaGo in 2016. Many people still just call it Zen though. As of 2017 it plays at strong professional level. In 2018 DeepZenGo retired from official matches.

Notable events:

  • March 2016. "DeepZenGo Project" launched.
  • November 2016. Lost an exhibition match 1-2 against Cho Chikun 9 dan. [ext] Usgo report.
  • Plays on Tygem and has beaten many professional players.
  • March 2017. AI participant in 1st World Go Championship. Beat #1 Japanese pro Iyama Yuta 9 dan, but lost against #1 Korean Park Junghwan 9 dan from a leading position and lost a close game against top Chinese 9 dan Mi Yuting.
  • June 2017. AI participant in 3rd MLILY Cup. Beat Shin Minjun 5 dan in the 1st round, but surprised many by losing to Wang Haoyang 6 dan in round 2.
  • June - December 2017. Playing a series of training games with the Japanese national training squad. After 100 games has ~80% win rate.
  • August 2017. Won 1st World AI Go Open. Lost to favourite FineArt and CGI in the preliminaries, but then beat them both in the knockout finals to claim the title (AlphaGo didn't participate).
  • December 2017 placed second after Golaxy in AI Ryusei 2017.
  • From 2017-06-12 to 2018-03-16 DeepZenGo played on Japanese platform Yuugen no Ma against professionals. In 3407 matches it won 3288 and lost 119 games reaching 96.5% winrate.
  • From 2018-03-24 to 2018-04-07. DeepZenGo retiring match (Denousen Final): 3 even games against strong pros. It lost to Mi Yuting, and won against Park Junghwan and Cho Chikun.

DeepZenGo last edited by SAlexander on November 6, 2020 - 00:37
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library