Leela (software)

Leela is a computer program for the Asian board game Go . It was programmed by the Belgian software developer Gian-Carlo Pascutto, who is also responsible for the powerful chess program Sjeng . At the International Computer Games Association's 13th Computer Olympics in 2008, Leela won a silver medal on the 9 × 9 board and a bronze medal on the 19 × 19 board.

The program is available for the Microsoft Windows , macOS and Linux operating systems ; the full version has been available on the Internet as freeware since April 2016 . In terms of software, it uses both Monte Carlo algorithms and, based on the AlphaGo program from DeepMind , deep learning technology based on an artificial neural network . To speed up the calculations, it can use the graphics processor of modern graphics cards via the OpenCL software interface . In addition to playing against Leela, the program can also be used for analysis, i.e. finding the best move in a position. It is possible to load and save games in SGF format .

With a skill level in the field of strong amateur players, Leela is one of the strongest free Go programs. On the KGS Go server , Leela achieved rank 8d in 2017 .

Leela Chess Zero

After DeepMind caused a sensation in December 2017 with a new publication on the generalized game engine AlphaZero , Gian-Carlo Pascutto and the American computer scientist Gary Linscott launched the open source project Leela Chess Zero , Lc0 for short , in a similar way to the previous one Leela Zero to replicate the results of DeepMind in chess and make them available to a wider group of users. At the end of 2018, the developers assumed that Lc0 would exceed Stockfish's skill level in the next few months . In May 2019, Lc0 won the Super Final of the Top Chess Engine Championship Season 15 against Stockfish. In the following tournament (season 16) Lc0 did not qualify for the super final; In 2020 the program won the title again.

Web links

Sjeng.org - Leela website of the program (English)

Lc0 for the Web ported browser version of Leela Chess Zero

As an autodidact to the top of the world The artificial chess intelligence Leela Chess Zero (Glarean Magazin April 2019)

Individual evidence

↑ https://arxiv.org/abs/1712.01815 Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm (December 5, 2017)
↑ https://www.chess.com/news/view/stockfish-wins-computer-chess-championship-rapid-lc0-finishes-3rd-3887 Stockfish Wins Computer Chess Championship Rapid; Lc0 Finishes 3rd (4th October 2018)
↑ https://www.newyorker.com/science/elements/how-the-artificial-intelligence-program-alphazero-mastered-its-games How the Artificial-Intelligence Program AlphaZero Mastered Its Games (December 28, 2018) ( with a few paragraphs on Lc0)
↑ A new age in computer chess? Lc0 beats Stockfish! Retrieved January 28, 2020 (English).

[1] ttps://arxiv.org/abs/1712.01815 Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm (December 5, 2017)

[2] ttps://www.chess.com/news/view/stockfish-wins-computer-chess-championship-rapid-lc0-finishes-3rd-3887 Stockfish Wins Computer Chess Championship Rapid; Lc0 Finishes 3rd (4th October 2018)

[3] ttps://www.newyorker.com/science/elements/how-the-artificial-intelligence-program-alphazero-mastered-its-games How the Artificial-Intelligence Program AlphaZero Mastered Its Games (December 28, 2018) ( with a few paragraphs on Lc0)

[4] A new age in computer chess? Lc0 beats Stockfish! Retrieved January 28, 2020 (English).