Package: wnpp Severity: wishlist Owner: Ximin Luo <infini...@debian.org>
* Package name : leela-zero Version : 0.15 Upstream Author : Gian-Carlo Pascutto <g...@sjeng.org> * URL : https://github.com/gcp/leela-zero * License : GPL-3 Programming Lang: C++ Description : Go engine with no human-provided knowledge, modeled after the AlphaGo Zero paper. Go engine with no human-provided knowledge, modeled after the AlphaGo Zero paper A Go program with no human provided knowledge. Using MCTS (but without Monte Carlo playouts) and a deep residual convolutional neural network stack. This is a fairly faithful reimplementation of the system described in the Alpha Go Zero paper "Mastering the Game of Go without Human Knowledge". For all intents and purposes, it is an open source AlphaGo Zero. https://deepmind.com/documents/119/agz_unformatted_nature.pdf No network weights are in this package. If you manage to obtain the AlphaGo Zero weights, this program will be about as strong, provided you also obtain a few Tensor Processing Units. Lacking those TPUs, the author recommends a top of the line GPU - it's not exactly the same, but the result would still be an engine that is far stronger than the top humans. Download the best known network weights file from: http://zero.sjeng.org/best-network