Difference between revisions of "Fat Fritz"

From Chessprogramming wiki
Jump to: navigation, search
Line 85: Line 85:
  
 
==Source Code==
 
==Source Code==
* [https://github.com/DanielUranga/lc0/tree/uci-wdl-reporting GitHub - DanielUranga/lc0 at uci-wdl-reporting]
+
* [https://github.com/DanielUranga/lc0/tree/uci-wdl-reporting GitHub - DanielUranga/lc0 at uci-wdl-reporting] by [[Daniel Uranga]]
 
* [https://ci.appveyor.com/project/DanielUranga/lc0 lc0 84 - AppVeyor]
 
* [https://ci.appveyor.com/project/DanielUranga/lc0 lc0 84 - AppVeyor]
 
* [https://github.com/kiudee/bayes-skopt GitHub - kiudee/bayes-skopt: A fully Bayesian implementation of sequential model-based optimization] by [[Karlson Pfannschmidt]] <ref>[https://en.chessbase.com/post/fat-fritz-update-and-fat-fritz-jr Fat Fritz 1.1 update and a small gift] by [[Albert Silver]]. [[ChessBase|ChessBase News]], March 05, 2020</ref>
 
* [https://github.com/kiudee/bayes-skopt GitHub - kiudee/bayes-skopt: A fully Bayesian implementation of sequential model-based optimization] by [[Karlson Pfannschmidt]] <ref>[https://en.chessbase.com/post/fat-fritz-update-and-fat-fritz-jr Fat Fritz 1.1 update and a small gift] by [[Albert Silver]]. [[ChessBase|ChessBase News]], March 05, 2020</ref>
 
* [https://github.com/yaneurao/YaneuraOu GitHub - yaneurao/YaneuraOu: YaneuraOu is the World's Strongest Shogi engine(AI player), WCSC29 1st winner, educational and USI compliant engine] » [[YaneuraOu]]
 
* [https://github.com/yaneurao/YaneuraOu GitHub - yaneurao/YaneuraOu: YaneuraOu is the World's Strongest Shogi engine(AI player), WCSC29 1st winner, educational and USI compliant engine] » [[YaneuraOu]]
 +
* [https://github.com/DanielUranga/Stockfish/commit/e44edde482ad06154c5133173245fdd2ad48a8cc#diff-a29c7f79728b32e998f606ae6f2fd2cb7c4a654fb0cef86547335542ca8ac054R38 ff2 · DanielUranga/Stockfish@e44edde · GitHub] by [[Daniel Uranga]]
  
 
==Rating Lists==
 
==Rating Lists==

Revision as of 22:59, 9 February 2021

Home * Engines * Fat Fritz

Unser Fritz [1]

Fat Fritz,
a commercial chess entity by ChessBase, the first version released in November 2019 featuring a set of custom made neural network weights that work in the open source project Leela Chess Zero within its Lc0 executable, while Fat Fritz 2.0 released in February 2021 is based on Stockfish 12 and custom NNUE technology.

Fat Fritz 1

Initially introduced as a cloud project in summer 2019, inspired by DeepMind's ground breaking AlphaZero approach combining Deep learning with Monte-Carlo Tree Search (MCTS) [2], and Albert Silver's Deus X experience, relaxing Leela Chess' "Zero" paradigm of pure reinforcement learning due to supervised learning, Fat Fritz was released in November 2019 as part the Fritz 17 package, best supported by a GPU card such as Nvidia GeForce 20 series. The project and its underlying techniques were emphasized by a talk of AlphaZero co-author Thore Graepel at ChessBase, attended by German computer chess icon Thomas Nitsche along with his sons and Math42 authors Raphael Nitsche and Maxim Nitsche [3]. Fat Fritz was further supported by Daniel Uranga [4] [5].

Lc0diagram.png
Deusx.png

Lc0 and DeusX aka Fat Fritz components [6]

Fat Fritz 2

Fat Fritz 2 is based on NNUE technology using Stockfish NNUE aka Stockfish 12 with a customized, double sized network, and thus, does not require GPU graphic cards for game playing [7]. The network was trained by Albert Silver using the original Fat Fritz, reinforced by Stockfish's alpha-beta search [8].

See also

Forum Posts

2019

2020

Re: Fat Fritz Update by Albert Silver, CCC, December 12, 2020

2021

External Links

Purchase

ChessBase News

2019

2020

2021

Fritz 17

Source Code

Rating Lists

References

Up one Level