AlphaGo: Mastering the ancient game of Go with Machine Learning

Posted by David Silver and Demis Hassabis, Google DeepMind
Games are a great testing ground for developing smarter, more flexible algorithms that have the ability to tackle problems in ways similar to humans. Creating programs that are able to play games better than the best humans has a long history - the first classic game mastered by a computer wasnoughts and crosses (also known as tic-tac-toe) in 1952 as a PhD candidate’s project. Then fellcheckers in 1994. Chess was tackled by Deep Blue in 1997. The success isn’t limited to board games, either - IBM's Watson won first place on Jeopardy in 2011, and in 2014 our own algorithms learned to play dozens of Atari games just from the raw pixel inputs.

But one game has thwarted A.I. research thus far: the ancient game of Go. Invented in China over 2500 years ago, Go is played by more than 40 million people worldwide. The rules are simple: players take turns to place black or white stones on a board, trying to capture the opponent's stones or surround empty space to make points of territory. Confucius wrote about the game, and its aesthetic beauty elevated it to one of the four essential arts required of any true Chinese scholar. The game is played primarily through intuition and feel, and because of its subtlety and intellectual depth it has captured the human imagination for centuries.

But as simple as the rules are, Go is a game of profound complexity. The search space in Go is vast -- more than a googol times larger than chess (a number greater than there are atoms in the universe!). As a result, traditional “brute force” AI methods -- which construct a search tree over all possible sequences of moves -- don’t have a chance in Go. To date, computers have played Go only as well as amateurs. Experts predicted it would be at least another 10 years until a computer could beat one of the world’s elite group of Go professionals.

We saw this as an irresistible challenge! We started building a system, AlphaGo, described in a paper in Nature this week, that would overcome these barriers. The key to AlphaGo is reducing the enormous search space to something more manageable. To do this, it combines a state-of-the-art tree search with two deep neural networks, each of which contains many layers with millions of neuron-like connections. One neural network, the “policy network”, predicts the next move, and is used to narrow the search to consider only the moves most likely to lead to a win. The other neural network, the “value network”, is then used to reduce the depth of the search tree -- estimating the winner in each position in place of searching all the way to the end of the game.

AlphaGo’s search algorithm is much more human-like than previous approaches. For example, when Deep Blue played chess, it searched by brute force over thousands of times more positions than AlphaGo. Instead, AlphaGo looks ahead by playing out the remainder of the game in its imagination, many times over - a technique known as Monte-Carlo tree search. But unlike previous Monte-Carlo programs, AlphaGo uses deep neural networks to guide its search. During each simulated game, the policy network suggests intelligent moves to play, while the value network astutely evaluates the position that is reached. Finally, AlphaGo chooses the move that is most successful in simulation.

We first trained the policy network on 30 million moves from games played by human experts, until it could predict the human move 57% of the time (the previous record before AlphaGo was 44%). But our goal is to beat the best human players, not just mimic them. To do this, AlphaGo learned to discover new strategies for itself, by playing thousands of games between its neural networks, and gradually improving them using a trial-and-error process known as reinforcement learning. This approach led to much better policy networks, so strong in fact that the raw neural network (immediately, without any tree search at all) can defeat state-of-the-art Go programs that build enormous search trees.

These policy networks were in turn used to train the value networks, again by reinforcement learning from games of self-play. These value networks can evaluate any Go position and estimate the eventual winner - a problem so hard it was believed to be impossible.

Of course, all of this requires a huge amount of compute power, so we made extensive use ofGoogle Cloud Platform, which enables researchers working on AI and Machine Learning to access elastic compute, storage and networking capacity on demand. In addition, new open source libraries for numerical computation using data flow graphs, such as TensorFlow, allow researchers to efficiently deploy the computation needed for deep learning algorithms across multiple CPUs or GPUs.

So how strong is AlphaGo? To answer this question, we played a tournament between AlphaGo and the best of the rest - the top Go programs at the forefront of A.I. research. Using a single machine, AlphaGo won all but one of its 500 games against these programs. In fact, AlphaGo even beat those programs after giving them 4 free moves headstart at the beginning of each game. A high-performance version of AlphaGo, distributed across many machines, was even stronger.

This figure from the Nature article shows the Elo rating and approximate rank of AlphaGo (both single machine and distributed versions), the European champion Fan Hui (a professional 2-dan), and the strongest other Go programs, evaluated over thousands of games. Pale pink bars show the performance of other programs when given a four move headstart.

It seemed that AlphaGo was ready for a greater challenge. So we invited the reigning 3-time European Go champion Fan Hui — an elite professional player who has devoted his life to Go since the age of 12 — to our London office for a challenge match. The match was played behind closed doors between October 5-9 last year. AlphaGo won by 5 games to 0 -- the first time a computer program has ever beaten a professional Go player.

AlphaGo’s next challenge will be to play the top Go player in the world over the last decade,Lee Sedol. The match will take place this March in Seoul, South Korea. Lee Sedol is excited to take on the challenge saying, "I am privileged to be the one to play, but I am confident that I can win." It should prove to be a fascinating contest!

We are thrilled to have mastered Go and thus achieved one of the grand challenges of AI. However, the most significant aspect of all this for us is that AlphaGo isn’t just an ‘expert’ system built with hand-crafted rules, but instead uses general machine learning techniques to allow it to improve itself, just by watching and playing games. While games are the perfect platform for developing and testing AI algorithms quickly and efficiently, ultimately we want to apply these techniques to important real-world problems. Because the methods we have used are general purpose, our hope is that one day they could be extended to help us address some of society’s toughest and most pressing problems, from climate modelling to complex disease analysis.

 
from:http://googleresearch.blogspot.com/2016/01/alphago-mastering-ancient-game-of-go.html

AlphaGo:用机器学习技术古老的围棋游戏掌握AlphaGo: Mastering the ancient game of Go with Machine Learning的更多相关文章

  1. 数据挖掘:实用机器学习技术P295页:

    数据挖掘:实用机器学习技术P295页: 在weka软件中的实验者界面中,新建好实验项目后,添加相应的实验数据,然后添加对应需要的分类算法 ,需要使用多个算法时候重复操作添加add algorithm. ...

  2. java围棋游戏源代码

    //李雨泽源代码,不可随意修改.//时间:2017年9月22号.//地点:北京周末约科技有限公司.//package com.bao; /*围棋*/ /*import java.awt.*; impo ...

  3. 谷歌发布"自动机器学习"技术 AI可自我创造

    谷歌发布"自动机器学习"技术 AI可自我创造 据Inverse报道,今年5月份,谷歌宣布其人工智能(AI)研究取得重大进展,似乎帮助科幻小说中最耸人听闻的末日预言成为现实.谷歌推出 ...

  4. 使用Java的GUI技术实现 “ 贪吃蛇 ” 游戏

    详细教程: 使用Java的GUI技术实现 " 贪吃蛇 " 游戏_IT打工酱的博客-CSDN博客

  5. 【机器学习Machine Learning】资料大全

    昨天总结了深度学习的资料,今天把机器学习的资料也总结一下(友情提示:有些网站需要"科学上网"^_^) 推荐几本好书: 1.Pattern Recognition and Machi ...

  6. 机器学习(Machine Learning)&深度学习(Deep Learning)资料【转】

    转自:机器学习(Machine Learning)&深度学习(Deep Learning)资料 <Brief History of Machine Learning> 介绍:这是一 ...

  7. 数据挖掘(data mining),机器学习(machine learning),和人工智能(AI)的区别是什么? 数据科学(data science)和商业分析(business analytics)之间有什么关系?

    本来我以为不需要解释这个问题的,到底数据挖掘(data mining),机器学习(machine learning),和人工智能(AI)有什么区别,但是前几天因为有个学弟问我,我想了想发现我竟然也回答 ...

  8. 机器学习笔记1 - Hello World In Machine Learning

    前言 Alpha Go在16年以4:1的战绩打败了李世石,17年又以3:0的战绩战胜了中国围棋天才柯洁,这真是科技界振奋人心的进步.伴随着媒体的大量宣传,此事变成了妇孺皆知的大事件.大家又开始激烈的讨 ...

  9. 学习笔记之机器学习(Machine Learning)

    机器学习 - 维基百科,自由的百科全书 https://zh.wikipedia.org/wiki/%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0 机器学习是人工智能的一个分 ...

随机推荐

  1. linux sed命令查询结果前后批量追加内容(html文件批量修改css,js等文件路径)

    1.需求:linux使用shell命令查询结果前后批量追加内容 例如:我需要在当前目录下所有的css文件路径前追加域名 我想的是用sed替换去实现,鲍哥的思路是用for循环 1.1方法1:鲍哥的for ...

  2. jmeter------线程组(默认)

    Jmeter中的采样器必须要基于线程组. 一.添加线程组 在测试计划上右键,然后选择,如下图: 二.线程组界面 三.线程组界面配置说明 1.名称:线程组自定义名称: 2.注释:添加的一些备注说明信息, ...

  3. Fresco 获得Bitmap

    ImageRequest imageRequest = ImageRequestBuilder .newBuilderWithSource( Uri.parse(getFeedItem(positio ...

  4. 一个简单的AboutMe页面

    Web2.0程序设计的小练习. Firefox下的效果 Chrome下的效果,套上了Helvetica,不知道为什么FF没有 (其实好像应该再优先加个Helvetica Neue的……呃……) 代码和 ...

  5. C++函数的重载、覆盖和隐藏区别

    a.成员函数被重载的特征:(1)相同的范围(在同一个类中)(2)函数名字相同(3)参数不同(4)virtual 关键字可有可无b.覆盖是指派生类函数覆盖基类函数,特征是:(1)不同的范围(分别位于派生 ...

  6. 【LOJ】#2280. 「FJOI2017」矩阵填数

    题解 我们发现没有限制的小方格可以随便填 然后考虑有限制的,我们把它切割成一个个小块(枚举相邻的横纵坐标),然后记录一下这个小块的最大值限制(也就是所有覆盖它的矩形最小的最大值) 记录一下每个小块的大 ...

  7. 六 Python基础 字符串和编码

    字符编码 我们已经讲过了,字符串也是一种数据类型,但是,字符串比较特殊的是还有一个编码问题. 因为计算机只能处理数字,如果要处理文本,就必须先把文本转换为数字才能处理.最早的计算机在设计时采用8个比特 ...

  8. React Native 系列(九)

    前言 本系列是基于React Native版本号0.44.3写的.很多的App都使用了Tab标签组件,例如QQ,微信等等,就是切换不同的选项,显示不同的内容.那么这篇文章将介绍RN中的Tab标签组件. ...

  9. Vue 2.0学习(四)计算属性

    {{}}模板内的表达式常用于简单的运算,当运算过长或逻辑复杂时,会难以维护. <div> {{ text.split(',').reverse().join('') }} </div ...

  10. Tsinsen Palisection

    建回文树. 正反建统计一种前缀和求出所有不相交的,用总数减去就是答案数. 在这里我们可以知道一个字符串中所有回文串的个数即为num数组之和(因为以一个节点为回文串结尾的字串都是唯一的) 也可以是cnt ...