From dfd938b73a242d21cb6692c9d7a677a0ab7b738e Mon Sep 17 00:00:00 2001 From: Range King Date: Fri, 27 Mar 2026 08:31:51 +0800 Subject: [PATCH] Add AlphaGo 2016-2017 example to READMEs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Insert a new 2016–2017 AlphaGo entry into README.md, README-ja.md, and README-zh.md describing AlphaGo's victories over Lee Sedol and Ke Jie. The additions emphasize that these achievements demonstrate a trained neural-network agent acquiring superhuman intuition, and are placed alongside existing historical examples (DQN, OpenAI Five, AlphaStar) for consistency across English, Japanese, and Chinese documentation. --- README-ja.md | 2 ++ README-zh.md | 2 ++ README.md | 2 ++ 3 files changed, 6 insertions(+) diff --git a/README-ja.md b/README-ja.md index 8717201b8..a9902dd39 100644 --- a/README-ja.md +++ b/README-ja.md @@ -18,6 +18,8 @@ Agent とはニューラルネットワークである -- Transformer、RNN、 - **2013 -- DeepMind DQN が Atari をプレイ。** 単一のニューラルネットワークが、生のピクセルとスコアだけを受け取り、7 つの Atari 2600 ゲームを学習 -- すべての先行アルゴリズムを超え、3 つで人間の専門家を打ち負かした。2015 年には同じアーキテクチャが [49 ゲームに拡張され、プロのテスターに匹敵](https://www.nature.com/articles/nature14236)、*Nature* に掲載。ゲーム固有のルールなし。決定木なし。一つのモデルが経験から学んだ。そのモデルが Agent だった。 +- **2016-2017 -- DeepMind AlphaGo が李世石と柯潔を破る。** 2016 年 3 月、AlphaGo はソウルで李世石に 4-1 で勝利 -- ニューラルネットワーク型の囲碁システムがフルマッチでトップ人間棋士を破った初の出来事だった。2017 年 5 月には烏鎮の Future of Go Summit で当時世界 1 位の柯潔を 3-0 で下した。人手で作った定石表が一手ごとに指示したわけではない。囲碁の名人が書いた記号ルールでもない。訓練されたモデルが、人類最古級の戦略ゲームの一つで超人的直観を獲得した。そのモデルが Agent だった。 + - **2019 -- OpenAI Five が Dota 2 を制覇。** 5 つのニューラルネットワークが 10 ヶ月間で [45,000 年分の Dota 2](https://openai.com/index/openai-five-defeats-dota-2-world-champions/) を自己対戦し、サンフランシスコのライブストリームで **OG** -- TI8 世界王者 -- を 2-0 で撃破。その後の公開アリーナでは 42,729 試合で勝率 99.4%。スクリプト化された戦略なし。メタプログラムされたチーム連携なし。モデルが完全に自己対戦を通じてチームワーク、戦術、リアルタイム適応を学んだ。 - **2019 -- DeepMind AlphaStar が StarCraft II をマスター。** AlphaStar は非公開戦で[プロ選手を 10-1 で撃破](https://deepmind.google/blog/alphastar-mastering-the-real-time-strategy-game-starcraft-ii/)、その後ヨーロッパサーバーで[グランドマスター到達](https://www.nature.com/articles/d41586-019-03298-6) -- 90,000 人中の上位 0.15%。不完全情報、リアルタイム判断、チェスや囲碁を遥かに凌駕する組合せ的行動空間を持つゲーム。Agent とは? モデルだ。訓練されたもの。スクリプトではない。 diff --git a/README-zh.md b/README-zh.md index fff83e48f..7f32702b3 100644 --- a/README-zh.md +++ b/README-zh.md @@ -18,6 +18,8 @@ Agent 是一个神经网络 -- Transformer、RNN、一个被训练出来的函 - **2013 -- DeepMind DQN 玩 Atari。** 一个神经网络,只接收原始像素和游戏分数,学会了 7 款 Atari 2600 游戏 -- 超越所有先前算法,在其中 3 款上击败人类专家。到 2015 年,同一架构扩展到 [49 款游戏,达到职业人类测试员水平](https://www.nature.com/articles/nature14236),论文发表在 *Nature*。没有游戏专属规则。没有决策树。一个模型,从经验中学习。那个模型就是 agent。 +- **2016-2017 -- DeepMind AlphaGo 击败李世石与柯洁。** 2016 年 3 月,AlphaGo 在首尔以 4-1 战胜李世石 -- 这是神经网络围棋系统首次在完整对局中击败顶尖人类棋手。2017 年 5 月,它又在乌镇未来围棋峰会上以 3-0 横扫当时世界排名第一的柯洁。不是人先把每一步棋都写好,再让它照着下。没有围棋名家写死的符号规则。一个训练出来的模型,在人类最古老的策略游戏之一中学出了超人的直觉。那个模型就是 agent。 + - **2019 -- OpenAI Five 征服 Dota 2。** 五个神经网络,在 10 个月内与自己对战了 [45,000 年的 Dota 2](https://openai.com/index/openai-five-defeats-dota-2-world-champions/),在旧金山直播赛上 2-0 击败了 **OG** -- TI8 世界冠军。随后的公开竞技场中,AI 在 42,729 场比赛中胜率 99.4%。没有脚本化的策略。没有元编程的团队协调逻辑。模型完全通过自我对弈学会了团队协作、战术和实时适应。 - **2019 -- DeepMind AlphaStar 制霸星际争霸 II。** AlphaStar 在闭门赛中 [10-1 击败职业选手](https://deepmind.google/blog/alphastar-mastering-the-real-time-strategy-game-starcraft-ii/),随后在欧洲服务器上达到[宗师段位](https://www.nature.com/articles/d41586-019-03298-6) -- 90,000 名玩家中的前 0.15%。一个信息不完全、实时决策、组合动作空间远超国际象棋和围棋的游戏。Agent 是什么?是模型。训练出来的。不是编出来的。 diff --git a/README.md b/README.md index c25a63458..682a6a8ed 100644 --- a/README.md +++ b/README.md @@ -17,6 +17,8 @@ The proof is written in history: - **2013 -- DeepMind DQN plays Atari.** A single neural network, receiving only raw pixels and game scores, learned to play 7 Atari 2600 games -- surpassing all prior algorithms and beating human experts on 3 of them. By 2015, the same architecture scaled to [49 games and matched professional human testers](https://www.nature.com/articles/nature14236), published in *Nature*. No game-specific rules. No decision trees. One model, learning from experience. That model was the agent. +- **2016-2017 -- DeepMind AlphaGo defeats Lee Sedol and Ke Jie.** In March 2016, AlphaGo beat Lee Sedol 4-1 in Seoul -- the first time a neural-network Go system defeated a top human player in a full match. In May 2017, it followed with a 3-0 sweep of world No. 1 Ke Jie at the Future of Go Summit in Wuzhen. No handcrafted opening book dictating every move. No symbolic rules written by Go masters. A trained model learned superhuman intuition in one of humanity's oldest strategy games. That model was the agent. + - **2019 -- OpenAI Five conquers Dota 2.** Five neural networks, having played [45,000 years of Dota 2](https://openai.com/index/openai-five-defeats-dota-2-world-champions/) against themselves in 10 months, defeated **OG** -- the reigning TI8 world champions -- 2-0 on a San Francisco livestream. In a subsequent public arena, the AI won 99.4% of 42,729 games against all comers. No scripted strategies. No meta-programmed team coordination. The models learned teamwork, tactics, and real-time adaptation entirely through self-play. - **2019 -- DeepMind AlphaStar masters StarCraft II.** AlphaStar [beat professional players 10-1](https://deepmind.google/blog/alphastar-mastering-the-real-time-strategy-game-starcraft-ii/) in a closed-door match, and later achieved [Grandmaster status](https://www.nature.com/articles/d41586-019-03298-6) on European servers -- top 0.15% of 90,000 players. A game with imperfect information, real-time decisions, and a combinatorial action space that dwarfs chess and Go. The agent? A model. Trained. Not scripted.