AlphaGo
AlphaGo mastered the ancient game of Go, defeated a Go world champion, and inspired a new era of AI systems. Making historyOur artificial intelligence (AI) system, AlphaGo, learned to master the ancient Chinese game of Go — a profoundly complex board game of strategy, creativity, and ingenuity. AlphaGo defeated a human Go world champion a decade before experts thought possible, inspired players ar
GoogleãŒå‡ºã—ãŸå›²ç¢ã‚½ãƒ•ãƒˆã€ŒAlphaGoã€ã®è«–文を翻訳ã—ã¦è§£èª¬ã—ã¦ã¿ã‚‹ã€‚ - 7rpn’s blog: ã†ã‚ã‚ã‚ã‚ã‚ãªæ—¥å¸¸
就活も無事終ã‚ã£ãŸã®ã§ï¼Œä¸€ç•ªã‚„ã‚ŠãŸã‹ã£ãŸAlphaGoã®è«–文を翻訳ã—ã¾ã—ãŸã€‚ ã”å˜çŸ¥ã®é€šã‚Šï¼Œé•·ã‚‰ã世界最強ã ã£ãŸå›²ç¢æ£‹å£«ã‚¤ãƒ»ã‚»ãƒ‰ãƒ«ä¹æ®µã‚’ç ´ã£ãŸGoogleã®å›²ç¢ãƒ—ãƒã‚°ãƒ©ãƒ ã§ã™ã€‚ è«–æ–‡ã®å†…容ã«è§¦ã‚Œã¤ã¤ä½•ã¨ãªã解説入れã¦ã„ããŸã„ã¨æ€ã„ã¾ã™ã€‚ãªã‚‹ã¹ã囲ç¢ã‚„DeepLearningを知らãªã„åˆå¿ƒè€…ã®äººã¨ã‹ã§ã‚‚仕組ã¿ã‚’ç†è§£ã§ãるよã†ã«åˆ†ã‹ã‚Šã‚„ã™ã書ã„ã¦ã„ã‘ã‚Œã°ã„ã„ãªã¨æ€ã„ã¾ã™ã€‚ 原題ã¯"Mastering the game of Go with deep neural networks and tree search"。 ã¨ã‚Šã‚ãˆãšæœ€åˆã®è¦ç´„ã®è¨³ã‹ã‚‰ã€‚ 謎ã®å˜èªžã¨ã‹ã¯å¾Œã‹ã‚‰èª¬æ˜Žå…¥ã‚Œã‚‹ã®ã§ï¼Œã•ã‚‰ã£ã¨æµã—èªã¿ã—ã¦ã„ãŸã ã‘ã‚Œã°ã€‚ 囲ç¢ã¯ã“ã‚Œã¾ã§AIã«ã¨ã£ã¦ã¨ã¦ã‚‚難ã—ã„ゲームã ã¨ã¿ãªã•ã‚Œã¦ããŸã€‚ãã‚Œã¯æŽ¢ç´¢ç¯„囲ãŒã¨ã¦ã‚‚広ã„ã“ã¨ã¨ï¼Œç›¤é¢ã®è©•ä¾¡ãŒé›£ã—ã„ãŸã‚。 ã“ã®è«–æ–‡ã§ã¯ï¼Œã‚³ãƒ³ãƒ”ュータを用ã„ãŸå›²ç¢ã®æ–°ã—ã„アプãƒãƒ¼
AlphaGo対æŽä¸–ドル - Wikipedia
本æ¥ã®è¡¨è¨˜ã¯ã€ŒAlphaGo対æŽä¸–ä¹ã€ã§ã™ã€‚ã“ã®è¨˜äº‹ã«ä»˜ã‘られãŸãƒšãƒ¼ã‚¸åã¯æŠ€è¡“çš„ãªåˆ¶é™ã¾ãŸã¯è¨˜äº‹åã®åˆ¶ç´„ã«ã‚ˆã‚Šä¸æ£ç¢ºãªã‚‚ã®ã¨ãªã£ã¦ã„ã¾ã™ã€‚ AlphaGo対æŽä¸–ä¹ï¼ˆã‚¢ãƒ«ãƒ•ã‚¡ã”・ãŸã„・イ・セドル）ã¯ã€éŸ“国ã®ãƒ—ãƒæ£‹å£«ã§ã‚ã‚‹æŽä¸–ä¹ã¨ã€Google DeepMindãŒé–‹ç™ºã—ãŸã‚³ãƒ³ãƒ”ュータ囲ç¢ãƒ—ãƒã‚°ãƒ©ãƒ AlphaGoã®äº”番å‹è² ã§ã‚る。 対局ã¯ã‚³ãƒŸ7ç›®åŠã®ä¸å›½ãƒ«ãƒ¼ãƒ«ã§ã€æŒã¡æ™‚é–“ã¯2時間ã§ã€åˆ‡ã‚Œã‚‹ã¨1手1分ã®ç§’èªã¿ã€ãŸã ã—1分å˜ä½ã§åˆè¨ˆ3回ã®è€ƒæ…®æ™‚é–“ãŒã‚ã‚‹[1]。 å‹è€…（å‹ã¡è¶Šã—者）ã¯100万米ドルã®è³žé‡‘を得る。もã—AlphaGoãŒå‹åˆ©ã™ã‚Œã°ã€è³žé‡‘ã¯UNICEFã‚’å«ã‚€ãƒãƒ£ãƒªãƒ†ã‚£ãƒ¼ã¸å¯„付ã•ã‚Œã‚‹[2]。賞金ã«åŠ ãˆã¦ã€æŽä¸–ä¹ã¯å…¨5戦ã®å¯¾å±€æ–™ã¨ã—ã¦15万米ドルã€1å‹ã«ã¤ã2万米ドルを得る[1]。 囲ç¢ã¯é•·ã„é–“ã€AIã®åˆ†é‡Žã«ãŠã‘る難å•ã¨è¦‹ãªã•ã‚Œã¦ãŠã‚Šã€ãƒã‚§ã‚¹ã‚„将棋ã§ã‚³ãƒ³ãƒ”ュータãŒãƒ—ãƒç›¸æ‰‹ã«å‹åˆ©ã‚’åŽã‚ã¦ã„ã‚‹ãªã‹ã€2
AlphaGo（アルファç¢ï¼‰ã¨ã‚¤ãƒ»ã‚»ãƒ‰ãƒ«ä¹æ®µã®å¯¾å±€ãŒæœ€çµ‚å±€ã¸ã€‚世界ä¸ãŒæ³¨ç›®ã™ã‚‹ã“ã®å¯¾å±€ã«ã¤ã„ã¦æŠ•ç¨¿ãŒé›†ã¾ã£ã¦ã„ã¾ã™ - 週刊ã¯ã¦ãªãƒ–ãƒã‚°
Google DeepMindã«ã‚ˆã£ã¦é–‹ç™ºã•ã‚ŒãŸå›²ç¢ãƒ—ãƒã‚°ãƒ©ãƒ 「AlphaGo（アルファç¢ï¼‰ã€ã¨ã€ãƒˆãƒƒãƒ—棋士ã§ã‚るイ・セドルä¹æ®µã®å¯¾å±€ãŒè©±é¡Œã¨ãªã£ã¦ã„ã¾ã™ã€‚ ã‚ã‚“ã¡ã¹ (id:AntiBayesian) ã•ã‚“ã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã§ã¯ã€Œãã‚‚ãã‚‚AlphaGoã¨ã¯ä½•ãªã®ã‹ã€ã€Œã‚¤ãƒ»ã‚»ãƒ‰ãƒ«ä¹æ®µã¯ã©ã‚“ãªæ£‹å£«ãªã®ã‹ã€ã¨ã„ã£ãŸè³ªå•ã«å¯¾ã™ã‚‹å›žç”ã‚’è¡Œã£ã¦ã„ã¾ã™ã€‚今ã‹ã‚‰ã§ã‚‚é…ããªã„ã§ã™ã€ã“ã¡ã‚‰ã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã‚’å‚考ã«ã“ã®ã€Œæ´å²çš„対局ã€ã«æ€ã„ã‚’ã¯ã›ã¦ã¿ã¾ã›ã‚“ã‹ï¼Ÿ antibayesian.hateblo.jp ã‚ã‚“ã¡ã¹ (id:AntiBayesian) ã•ã‚“ã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã«ã‚‚ã‚るよã†ã«ã€å›²ç¢ã¯ã€ãƒã‚§ã‚¹ãªã©ä»–ã®ãƒœãƒ¼ãƒ‰ã‚²ãƒ¼ãƒ ã¨æ¯”較ã™ã‚‹ã¨ã‚³ãƒ³ãƒ”ュータãŒäººé–“ã«å‹ã¤ã®ãŒå›°é›£ã§ã‚ã‚‹ã¨è€ƒãˆã‚‰ã‚Œã¦ã„ã¾ã—ãŸã€‚id:shi3z ã•ã‚“ã®ã‚¨ãƒ³ãƒˆãƒªãƒ¼ã§ã¯AlphaGoã®ç‰¹å¾´ã«ã¤ã„ã¦è§£èª¬ãŒãªã•ã‚Œã¦ã„ã¾ã™ã€‚ http://d.hatena.ne.jp/
New community features for Google Chat and an update on Currents
Join the official community for Google Workspace administrators In the Google Cloud Community, connect with Googlers and other Google Workspace admins like yourself. Participate in product discussions, check out the Community Articles, and learn tips and tricks that will make your work and life easier. Be the first to know what's happening with Google Workspace. ______________ Learn about more Goo
囲ç¢ãƒãƒ£ãƒ³ãƒ”オンを打ã¡ç ´ã£ãŸGoogleã®äººå·¥çŸ¥èƒ½ã€ŒAlphaGoã€ã‚’作ã£ãŸå¤©æ‰ãƒ‡ãƒŸã‚¹ãƒ»ãƒã‚µãƒ“スãŒäººå·¥çŸ¥èƒ½ã‚’語る
GoogleãŒ500億円以上ã§è²·åŽã—ãŸè¬Žã®ã‚¹ã‚¿ãƒ¼ãƒˆã‚¢ãƒƒãƒ—「DeepMindã€ã€‚当時ã€ç„¡åã®äººå·¥çŸ¥èƒ½(AI)開発ベンãƒãƒ£ãƒ¼ã®DeepMindã‚’GoogleãŒè²·åŽã—ãŸç›®çš„ã¯ã€å¤©æ‰ãƒ‡ãƒŸã‚¹ãƒ»ãƒã‚µãƒ“スæ°ã‚’手ä¸ã«åŽã‚ãŸã„ã‹ã‚‰ã¨å™‚ã•ã‚Œã¦ã„ã¾ã—ãŸã€‚Googleã®æœŸå¾…通りDeepMindã¯AI技術を磨ã上ã’ã€å¦ç¿’ã—ã¦ã‚²ãƒ¼ãƒ ã®è…•ã‚’メã‚メã‚上ã’るアルゴリズム「DQNã€ã‚’発表ã—ã¦ä¸€èºæœ‰åã«ãªã‚Šã€äººå·¥çŸ¥èƒ½ã‚½ãƒ•ãƒˆã€ŒAlphaGoã€ã‚’開発ã—ã¦ã“ã‚Œã¾ã§é›£æ”»ä¸è½ã¨è€ƒãˆã‚‰ã‚Œã¦ããŸå›²ç¢ã®ä¸–ç•Œæœ€å¼·æ£‹å£«ã‚’ç ´ã‚‹ã®ã«æˆåŠŸã™ã‚‹ãªã©ã€ä¸–é–“ã‚’ã‚ã£ã¨é©šã‹ã›ã¦ã„ã¾ã™ã€‚ ãã®DeepMindを率ã„ã‚‹ãƒã‚µãƒ“スæ°ã«The VergeãŒã€AlphaGo VS 囲ç¢ç•Œã®é”王イ・セドル(æŽä¸–ä¹)ä¹æ®µã®ä¸–ç´€ã®å¯¾æ±ºã®ç¬¬1戦ãŒçµ‚了ã—ãŸç¿Œæ—¥ã®ç¬¬2戦ãŒå§‹ã¾ã‚‹å‰ã«ã‚¤ãƒ³ã‚¿ãƒ“ューã—ã¦ã„ã¾ã™ã€‚ DeepMind founder Demis Hassabis on how A
AlphaGoã¨ãƒˆãƒƒãƒ—棋士ã®å¯¾å±€ã‚’観戦ã—よã†ï¼ - ã‚ã‚“ã¡ã¹ï¼
"We cannot go against the trend. I think machines will beat humans someday. " "If I get defeated it might be negative for go and this is inevitable in this modern life. However, it will not destroy the value of go itself." （「ç§ãŸã¡ã¯æ™‚代ã®æµã‚Œã«é€†ã‚‰ã†ã“ã¨ã¯å‡ºæ¥ã¾ã›ã‚“。ã„ã¤ã‹æ©Ÿæ¢°ã¯äººé–“を打ã¡è² ã‹ã™ã§ã—ょã†ã€ 「もã—ç§ãŒè² ã‘ãŸã‚‰å›²ç¢ã«ã¨ã£ã¦ã¯ãƒã‚¬ãƒ†ã‚£ãƒ–ãªã“ã¨ã§ã™ãŒã€ã“ã‚Œã¯ç¾ä»£ç¤¾ä¼šã§ã¯é¿ã‘られãªã„ã“ã¨ã§ã™ã€‚ ã—ã‹ã—ã€ãã®æ•—北ã¯å›²ç¢è‡ªèº«ã®ä¾¡å€¤ã‚’æãªã†ã“ã¨ã¯ãªã„ã§ã—ょã†ã€ã‚¤ãƒ»ã‚»ãƒ‰ãƒ«ï¼‰ Human-AI showdown begins 1 p.m. 2016å¹´3月9æ—¥ã€"Alph
＜囲ç¢ï¼šäººé–“ｖｓ人工知能＞神秘ã®é ˜åŸŸã€ä¸å¤®ã®ã€ŒåŽšã¿ã€â€¦ã‚¢ãƒ«ãƒ•ã‚¡ç¢ã¯è¨ˆç®—ã—ãŸ
アルファç¢ãŒï¼•ï¼ï¼ï¼å¹´é–“続ã„ã¦ããŸå›²ç¢ã®åŽŸç†ã‚’æ ¹æœ¬ã‹ã‚‰æ›¸ãæ›ãˆã¤ã¤ã‚ã‚‹ã€‚æ ¸å¿ƒã¯ä¸å¤®æ”»ç•¥ã 。ã‹ã¤ã¦äººé–“ãŒã€ŒåŽšã¿ã€ã¨å‘½åã—ã¦ç¥žç§˜ã®é ˜åŸŸã¨ã—ã¦æ®‹ã—ã¦ããŸç©ºé–“ã‚’ã€ã‚¢ãƒ«ãƒ•ã‚¡ç¢ã¯ã¤ã„ã«ç²¾å¯†ãªè¨ˆç®—力ã§å¾æœã—é‚ã’ã¦ã„る。 第２局ã¯ç¬¬ï¼‘å±€ã¨ã¯é•ã£ãŸã€‚対局ä¸ãšã£ã¨æŽä¸–ドル（イ・セドル）ä¹æ®µã¯å†·é™ã‹ã¤æŸ”軟ã€æ™‚ã«ã¯æžœæ•¢ã ã£ãŸã€‚特別ã«å¤±ç€ï¼ˆèª¤ã£ãŸæ‰‹ã‚’打ã¤ã“ã¨ï¼‰ã‚‚ãªã‹ã£ãŸã€‚ã‚€ã—ã‚アルファç¢ãŒç„¡ç†ãªæ–¹æ³•ã¨å”çªãªæ‰‹ã‚’連発ã—ãŸã€‚終盤ã®æŒ¯ã‚Šæ›¿ã‚り（フリカワリ）部分ã¯ã€å¤§ããªéŒ¯èª¤ã®ã‚ˆã†ã«ã¿ãˆãŸã€‚ã ãŒã€ãã‚Œã¯äººé–“ã®ç›®ã‹ã‚‰è¦‹ãŸæ™‚ã§ã‚ã£ãŸã€‚人工知能ã®æ–‡æ³•ã¯é•ã£ã¦ã„ãŸã€‚ ä¸åŽŸï¼ˆç¬¬ï¼•ç·šã‹ã‚‰ä¸å¤®ã‚ãŸã‚Šï¼‰ã®é‹å–¶åŠ›ã«ãŠã„ã¦ã‚¢ãƒ«ãƒ•ã‚¡ç¢ã¯å„ªç§€ã ã£ãŸã€‚åºç›¤ãƒ»ä¸ç›¤ã«ã‚¢ãƒ«ãƒ•ã‚¡ç¢ãŒå³è¾ºã«ï¼“ï¼—ã‚’ç½®ã„ãŸæ™‚ã€è¦³æˆ¦å®¤ã¯ã–ã‚ã¤ã„ãŸã€‚イ・ヒソンä¹æ®µã¯ã€Œæœ¬ã«ã¯ãªã„手ã€ã¨è¨€ã£ãŸã€‚アルファç¢ãŒç¬¬ï¼•ç·šã§è‚©ãƒ„ã‚（相手ã®çŸ³ã®æ–œã‚上ã«æ‰“ã¤æ‰‹ï¼‰ã—ãŸãŸã‚ã 。 通常ã€å›²ç¢ã«ãŠã„ã¦ç¬¬ï¼“ç·š
1
リリースã€éšœå®³æƒ…å ±ãªã©ã®ã‚µãƒ¼ãƒ“スã®ãŠçŸ¥ã‚‰ã›
最新ã®äººæ°—エントリーã®é…ä¿¡
j次ã®ãƒ–ックマーク
kå‰ã®ãƒ–ックマーク
lã‚ã¨ã§èªã‚€
eコメント一覧を開ã
oページを開ã
入門AlphaGo
cakes（ケイクス）
cakes（ケイクス）
長文日記
{{#tags}}- {{label}}
{{/tags}}