李根唐旭發(fā)自凹非寺量子位報(bào)道|公眾號(hào)QbitAI3月27日，2018年圖靈獎(jiǎng)嘉獎(jiǎng)?wù)浇視裕荷疃葘W(xué)習(xí)三巨頭YoshuaBengio、GeoffreyHinton，YannLeCun一起榮膺計(jì)算機(jī)領(lǐng)域的最高榮譽(yù)。ACM評(píng)委會(huì)頒獎(jiǎng)詞稱：表彰他們以

李根 唐旭 發(fā)自 凹非寺

量子位 報(bào)道 | 公眾號(hào) QbitAI

3月27日，2018年圖靈獎(jiǎng)嘉獎(jiǎng)?wù)浇視裕荷疃葘W(xué)習(xí)三巨頭Yoshua Bengio、Geoffrey Hinton，Yann LeCun一起榮膺計(jì)算機(jī)領(lǐng)域的最高榮譽(yù)。

ACM評(píng)委會(huì)頒獎(jiǎng)詞稱：表彰他們以概念和工程的突破，讓深度神經(jīng)網(wǎng)絡(luò)成為計(jì)算關(guān)鍵部件。

其中，72歲的Geoffrey Hinton更是眾望所歸，在年盛一年的呼聲之后，終于加冕圖靈獎(jiǎng)。

評(píng)選揭曉后，量子位獲悉，其實(shí)今年全球多位AI領(lǐng)域影響力科學(xué)家，就已經(jīng)提名Geoffrey Hinton且撰寫了推薦信。

按照?qǐng)D靈獎(jiǎng)提名推薦流程，推薦人須與被推薦者曾經(jīng)共事。

所以包括Google研究負(fù)責(zé)人Jeff Dean、創(chuàng)新工場(chǎng)董事長(zhǎng)李開復(fù)、微軟研究院掌舵者Eric Horvitz，以及此次一同獲獎(jiǎng)的Yann LeCun等，都因曾經(jīng)共事而提名了Geoffery Hinton。

現(xiàn)在，我們獲得授權(quán)，披露李開復(fù)向ACM圖靈獎(jiǎng)評(píng)選委員會(huì)提名Geoff Hinton的推薦信。

需要說明的是，原件為英文，量子位進(jìn)行了編譯。

但強(qiáng)烈推薦學(xué)有余力的盆友直接讀英文原文(后附)，一方面是因?yàn)槔铋_復(fù)圍繞學(xué)術(shù)、產(chǎn)業(yè)和教育等三大方面，對(duì)Geoffrey Hinton進(jìn)行功績(jī)介紹，字里行間也不乏一些鮮為人知的小故事，還有溢于言表的欽佩和贊美。

另一方面，也確實(shí)鍛煉英文的好機(jī)會(huì)。

OK，Here we go~

李開復(fù)推薦信（量子位編譯版）

尊敬的ACM圖靈獎(jiǎng)評(píng)選委員會(huì)：

我謹(jǐn)以此文表達(dá)個(gè)人最強(qiáng)烈的推薦和支持，提名Geoff Hinton為本年度的圖靈獎(jiǎng)候選人。這是人工智能的時(shí)代，而在人工智能領(lǐng)域，沒人能比Geoff更有資格獲得這份嘉獎(jiǎng)。

我現(xiàn)在是創(chuàng)新工場(chǎng)董事長(zhǎng)及CEO，并曾作為高管任職于蘋果、SGI、微軟和谷歌。我曾是卡內(nèi)基梅隆大學(xué)的助理教授，同時(shí)在這所大學(xué)獲得了博士學(xué)位。

也正是在CMU，1983年，我作為一名博士生結(jié)識(shí)了Geoff：修讀了他的神經(jīng)網(wǎng)絡(luò)課程；并和他的研究團(tuán)隊(duì)一起，嘗試將神經(jīng)網(wǎng)絡(luò)應(yīng)用于語音識(shí)別；我還在他的指導(dǎo)下完成了自己的輔修課程論文（關(guān)于將貝葉斯學(xué)習(xí)應(yīng)用于博弈游戲），我的博士論文（用機(jī)器學(xué)習(xí)方法進(jìn)行語音識(shí)別），也曾向Geoff和他的團(tuán)隊(duì)尋求過建議。

△Hinton和李開復(fù)2017年會(huì)面，量子位配圖

雖然Geoff并非我的博士生導(dǎo)師，但他對(duì)于我博士論文的影響卻十分巨大。他的學(xué)生Peter Brown（統(tǒng)計(jì)機(jī)器翻譯的共同發(fā)明人，如今文藝復(fù)興科技的CEO）當(dāng)時(shí)是我的老師，也正是他教會(huì)了我如何把不同種類的機(jī)器學(xué)習(xí)算法應(yīng)用于語音識(shí)別，為我的博士論文打下了基礎(chǔ)。

其后1988年，我的博士論文實(shí)現(xiàn)了當(dāng)時(shí)最好的語音識(shí)別模型，也促使語音識(shí)別領(lǐng)域，將重心從專家系統(tǒng)方法轉(zhuǎn)移到機(jī)器學(xué)習(xí)方法。鑒于Geoff的才華、堅(jiān)韌和慷慨，如果我能從Geoff和Peter身上受益如此之多，想必還有成千上萬的受益者跟我一樣。

Geoff對(duì)于AI領(lǐng)域的貢獻(xiàn)是巨大、無可比擬的。在我最近出版的暢銷書 AI Supowerpowers: China, Silicon Valley, and the New World Order（中文名《AI·未來》）中，以通俗的說法描述了Geoff對(duì)于AI領(lǐng)域的貢獻(xiàn)：

曾經(jīng)，人工神經(jīng)網(wǎng)絡(luò)能做的事非常有限。復(fù)雜問題若要得出準(zhǔn)確的結(jié)果，必須構(gòu)建很多層的人工神經(jīng)元，但神經(jīng)元層數(shù)增加后，研究人員當(dāng)時(shí)還未找到針對(duì)新增神經(jīng)元有效的訓(xùn)練方法。

在21世紀(jì)頭10年的中期，深度學(xué)習(xí)這項(xiàng)重大的技術(shù)性突破終于到來，知名研究人員杰弗里·辛頓找到了有效訓(xùn)練人工神經(jīng)網(wǎng)絡(luò)中新增神經(jīng)元層的方法。這就像是給舊的神經(jīng)網(wǎng)絡(luò)打了一針興奮劑，使它們的能力倍增，可以執(zhí)行更多、更復(fù)雜的工作，例如語音及物體識(shí)別。

性能大增的人工神經(jīng)網(wǎng)絡(luò)——現(xiàn)在有了新的名字“深度學(xué)習(xí)”——開始把舊的系統(tǒng)遠(yuǎn)遠(yuǎn)甩在身后。多年來對(duì)神經(jīng)網(wǎng)絡(luò)根深蒂固的成見讓人工智能的許多研究人員忽略了這個(gè)已經(jīng)取得出色成果的“邊緣群體”，但2012年杰弗里·辛頓的團(tuán)隊(duì)在一場(chǎng)國(guó)際計(jì)算機(jī)視覺競(jìng)賽中的勝出[1]，讓人工神經(jīng)網(wǎng)絡(luò)和深度學(xué)習(xí)重新回到聚光燈下。

在邊緣地帶煎熬了數(shù)十年后，以深度學(xué)習(xí)的形式再次回到公眾視野中的神經(jīng)網(wǎng)絡(luò)法不僅成功地讓人工智能回暖，也第一次把人工智能真正地應(yīng)用在現(xiàn)實(shí)世界中。研究人員、未來學(xué)家、科技公司CEO都開始討論人工智能的巨大潛力：識(shí)別人類語言、翻譯文件、識(shí)別圖像、預(yù)測(cè)消費(fèi)者行為、辨別欺詐行為、批準(zhǔn)貸款、幫助機(jī)器人“看”，甚至開車。

如果就學(xué)術(shù)成果而言，Geoff的引用超過25萬次，其中一半以上來自于過去5年；他的H指數(shù)是驚人的142；他是波茨曼機(jī)以及通過梯度下降法實(shí)現(xiàn)反向傳播（前者與Terry Sejnowski共同發(fā)布于1983年，后者則在1986年與David Rumelhart共同發(fā)布于《自然》雜志）兩項(xiàng)開創(chuàng)性工作的共同發(fā)明人，這些工作引入了神經(jīng)網(wǎng)絡(luò)中隱藏層的思想，以及一種用于對(duì)其附屬參數(shù)進(jìn)行訓(xùn)練的優(yōu)美、易于計(jì)算的方法。

隱藏層讓軟件從“人類控制“（如專家系統(tǒng)）下得到解放，而反向傳播則讓非線性組合系統(tǒng)發(fā)現(xiàn)突出特征成為可能，以一種更接近于目標(biāo)而非人類驅(qū)動(dòng)的方式。

然而，這些思想對(duì)于其所處時(shí)代而言，當(dāng)時(shí)過于超前。因?yàn)楫?dāng)時(shí)并沒有足夠的數(shù)據(jù)和算力，能讓這些理論上的方法解決實(shí)際生活中的問題，或是在競(jìng)爭(zhēng)中壓倒其他方法。

于是在20世紀(jì)80年代初期，該領(lǐng)域被專家系統(tǒng)所統(tǒng)治，直到80年代末期，專家系統(tǒng)因過于脆弱和難以擴(kuò)展成為歷史。

不過替代專家系統(tǒng)的也并非Geoff提出的構(gòu)想（那時(shí)還太早），而是妥協(xié)于少量數(shù)據(jù)和算力的簡(jiǎn)化版本的神經(jīng)網(wǎng)絡(luò)。

我的博士論文（使用了隱馬爾科夫模型）討論的就是其中的一種。這些簡(jiǎn)化過的方法確實(shí)能在一部分應(yīng)用上作出貢獻(xiàn)，但如同專家系統(tǒng)一樣，它們并不能在那些最難解決的問題上進(jìn)行擴(kuò)展，如下圍棋、人類水平的語音和視覺。

然而時(shí)來運(yùn)轉(zhuǎn)，從1985年到2015年，全球數(shù)據(jù)量和算力發(fā)生了巨大的增長(zhǎng)。

舉例來說，我在1988年的博士論文使用了當(dāng)時(shí)最大的語音數(shù)據(jù)庫，但它也只有100MB的大小。

而今天，最好的語音識(shí)別系統(tǒng)要在100TB的數(shù)據(jù)上進(jìn)行訓(xùn)練——這是一百萬倍的提升。有了數(shù)據(jù)上如此巨大的提升作為支持，Geoff的思想最終閃耀了起來——他的方法能將層的數(shù)量由1增加到1000，而數(shù)據(jù)量和模型復(fù)雜度的提升會(huì)使得深度學(xué)習(xí)系統(tǒng)持續(xù)進(jìn)步。

事后再看這些問題，顯得異常容易。但在當(dāng)時(shí)，現(xiàn)實(shí)卻是非常殘酷的。20世紀(jì)90年代對(duì)于像Geoff一樣的神經(jīng)網(wǎng)絡(luò)研究者而言，是最黑暗的時(shí)刻。

Geoff早期的理論工作創(chuàng)造出了智能的火花，但數(shù)據(jù)和算力的匱乏卻阻礙了這些深度學(xué)習(xí)系統(tǒng)展示出更優(yōu)秀的性能。隨著科研經(jīng)費(fèi)消耗殆盡，許多神經(jīng)網(wǎng)絡(luò)研究者將自己的工作轉(zhuǎn)移到了其他領(lǐng)域。然而，面對(duì)黯淡而又浮躁的科研資助環(huán)境，Geoff依然作為少數(shù)研究者（其他關(guān)鍵性研究者包括Yann LeCun和Yoshua Bengio）堅(jiān)持了下來，不懈地將神經(jīng)網(wǎng)絡(luò)方法繼續(xù)向前推進(jìn)。

他搬到了加拿大，根據(jù)受限的經(jīng)費(fèi)環(huán)境調(diào)整了團(tuán)隊(duì)，而后繼續(xù)努力將科學(xué)的邊界向前拓展。

Geoff在接下來的30年中持續(xù)地為神經(jīng)網(wǎng)絡(luò)方法做著貢獻(xiàn)，其中包括多專家模型、亥姆霍茲?rùn)C(jī)、神經(jīng)動(dòng)畫生成系統(tǒng)、概率推理模型、隨機(jī)領(lǐng)域嵌入、鄰域組件分析、t-SNE，以及諸多創(chuàng)新思想和研究。

鮮有某項(xiàng)技術(shù)可以徹底顛覆多個(gè)領(lǐng)域的研究，而深度學(xué)習(xí)就是其中之一。從2010年到2016年，本質(zhì)上講，整個(gè)感知領(lǐng)域——語音識(shí)別、圖像識(shí)別、計(jì)算機(jī)視覺都轉(zhuǎn)移到了深度學(xué)習(xí)的路徑上，就是因?yàn)镚eoff和他的同事們證明了——對(duì)于感知而言，深度學(xué)習(xí)就是最佳也最能普及的方法。

在整個(gè)人工智能領(lǐng)域，人類的感知能力（聽、看和理解）被視作人類獨(dú)有的天賦，但對(duì)于AI而言，這是一個(gè)巨大挑戰(zhàn)。值得一提的是，還有一項(xiàng)巨大的挑戰(zhàn)是下圍棋，已經(jīng)被Deepmind開發(fā)的AlphaGo，同樣使用深度學(xué)習(xí)方法攻克了。當(dāng)時(shí)震驚了整個(gè)世界，也成了“AI革命”的催化劑。

現(xiàn)在回顧，Geoff的團(tuán)隊(duì)是如此顛覆計(jì)算機(jī)視覺研究的：2012年，他的團(tuán)隊(duì)構(gòu)建了一個(gè)基于神經(jīng)網(wǎng)絡(luò)的系統(tǒng)，在ImageNet 1000個(gè)類的物體識(shí)別競(jìng)賽中將錯(cuò)誤率一下降低了40%。

在此之前，計(jì)算機(jī)視覺領(lǐng)域的研究社群已經(jīng)習(xí)慣了每年一小部分的增量提升，而Geoff團(tuán)隊(duì)的成績(jī)震驚了整個(gè)社群——人們未曾想象過，一個(gè)“局外人”會(huì)以一種“非傳統(tǒng)方法”以如此大的優(yōu)勢(shì)贏下競(jìng)賽。

如果說反向傳播是Geoff最重要的理論貢獻(xiàn)，那么其團(tuán)隊(duì)在ImageNet競(jìng)賽中取得的成果則是Geoff最為人認(rèn)可的貢獻(xiàn)。那屆ImageNet競(jìng)賽結(jié)果所掀起的微波，最終成為了深度學(xué)習(xí)大潮中的滔天巨浪。

深度學(xué)習(xí)的浪潮如今也正在改變每一個(gè)行業(yè)。舉例來說，作為一名身處中國(guó)的投資人，我本人也品嘗過這股春風(fēng)帶來的甘霖：Geoff在2012年發(fā)表的論文以及當(dāng)年ImageNet上的成果，為4家中國(guó)的計(jì)算機(jī)視覺公司帶來了靈感，而現(xiàn)在他們的總估值超過120億美元。但請(qǐng)記住，這還只是Geoff的工作在一個(gè)國(guó)家、一個(gè)小小的領(lǐng)域內(nèi)帶來的成果。

此外，Geoff的工作還使得深度學(xué)習(xí)顛覆了語音識(shí)別領(lǐng)域（也是我博士時(shí)期的研究方向），幫助當(dāng)時(shí)加盟百度的吳恩達(dá)，在2015年使得機(jī)器識(shí)別的準(zhǔn)確度超越人類。

從更廣闊的視角上看，世界上的每一家科技巨頭（谷歌、微軟、IBM、Facebook、亞馬遜、百度、騰訊和阿里巴巴）都為深度學(xué)習(xí)打造了自己的平臺(tái)，甚至將自己重新標(biāo)榜為“AI公司”；而在風(fēng)投領(lǐng)域，我們則見證了深度學(xué)習(xí)驅(qū)動(dòng)下大批獨(dú)角獸公司的出現(xiàn)（僅在中國(guó)就有超過20家）。

同時(shí)，由于深度學(xué)習(xí)需要的強(qiáng)大算力無法從傳統(tǒng)的CPU中獲得，為了負(fù)載深度學(xué)習(xí)所需的工作量，GPU開始被大規(guī)模使用，并由此引發(fā)了英偉達(dá)的崛起以及半導(dǎo)體工業(yè)的復(fù)活。

最為重要的是，我們的生活已經(jīng)發(fā)生了深刻的變化：從搜索引擎，到社交網(wǎng)絡(luò)再到電子商務(wù)，從無人商店到無人汽車，從金融到醫(yī)療，幾乎所有能想象到的領(lǐng)域，都被深度學(xué)習(xí)的力量所重塑，或是顛覆。

在任何具有充足數(shù)據(jù)的領(lǐng)域，深度學(xué)習(xí)在用戶體驗(yàn)、用戶粘性、營(yíng)收和利潤(rùn)方面都帶來了極大的提升。深度學(xué)習(xí)背后的核心思想（始于反向傳播）——一個(gè)目標(biāo)函數(shù)可以被用來使商業(yè)指標(biāo)最大化——已經(jīng)對(duì)所有行業(yè)造成了深刻的影響，并幫助那些擁有數(shù)據(jù)、擁抱機(jī)器學(xué)習(xí)的公司獲得了難以置信的利潤(rùn)。

總的說來，人工智能可以說是當(dāng)今我們能夠成熟應(yīng)用的技術(shù)中，最為令人興奮的一種。普華永道和麥肯錫預(yù)計(jì)，在2030年以前，AI會(huì)給全球GDP帶來12到16萬億美元的增長(zhǎng)。

而AI領(lǐng)域最重要的進(jìn)展，以及AI技術(shù)的成熟度被堅(jiān)信的首要原因，就是Geoff在深度學(xué)習(xí)方面的工作成果。

誠(chéng)然，每一位圖靈獎(jiǎng)的獲獎(jiǎng)?wù)叨紝?duì)計(jì)算機(jī)科學(xué)領(lǐng)域有著極其重大的影響，但極少數(shù)能像Geoff一樣，改變了整個(gè)世界。

在變革者的角色之外，Geoff還是一位真正的思想領(lǐng)袖。雖然他總是言辭溫和，他卻是一位真正塑造并重塑整個(gè)研究社群的精神領(lǐng)袖。

他不知疲倦地教誨，不只是對(duì)他的學(xué)生，更是對(duì)這個(gè)世界。

正如他1986年創(chuàng)辦聯(lián)結(jié)主義學(xué)院時(shí)，親自與在計(jì)算機(jī)視覺和語音處理領(lǐng)域的人們?nèi)ミM(jìn)行溝通，說服他們?nèi)ダ斫獠肀C(jī)器學(xué)習(xí)。然而，當(dāng)2018年所有的工作都獲得了成功，整個(gè)世界都投入了深度學(xué)習(xí)的懷抱之時(shí)，他還繼續(xù)指出一條全新的道路。

在行業(yè)紛紛向深度學(xué)習(xí)靠攏，大公司們不斷收集更多的數(shù)據(jù)并開始引領(lǐng)深度學(xué)習(xí)的“工業(yè)化”之時(shí)，他卻號(hào)召人們向前一步，去創(chuàng)造“下一個(gè)深度學(xué)習(xí)”。換而言之，解決AI根本問題的一種全新方法，幫助機(jī)器更接近真正的人類智慧。

他在思想上的領(lǐng)袖魅力來源于他畢生的愿景，以及對(duì)于更好地去理解人類認(rèn)知能力的追尋。盡管深度學(xué)習(xí)是一項(xiàng)正在改變世界的重大突破，他卻僅僅將其視為實(shí)現(xiàn)自己長(zhǎng)期愿景路途中的一塊踏腳石。他最近在膠囊網(wǎng)絡(luò)方面的新工作，也正再一次讓研究者們重新審視自身在Geoff愿景中的角色和責(zé)任。

我堅(jiān)信，Geoff就是今天的人工智能領(lǐng)域內(nèi)最重要的人物，沒有之一。

他對(duì)于學(xué)界和業(yè)界的貢獻(xiàn)同樣地突出。他不僅是一位優(yōu)秀的、引領(lǐng)性的學(xué)者，亦是一位孜孜以求、慷慨、堅(jiān)韌、優(yōu)雅、有原則的紳士。他是所有年輕計(jì)算機(jī)科學(xué)家的楷模。他的工作大大超越了神經(jīng)網(wǎng)絡(luò)和機(jī)器學(xué)習(xí)的范疇，極大地影響了計(jì)算機(jī)視覺、語音及信號(hào)處理、統(tǒng)計(jì)學(xué)、認(rèn)知科學(xué)以及神經(jīng)科學(xué)。

我想不出其他任何人比他更有資格獲得圖靈獎(jiǎng)，并在此敦促評(píng)選委員會(huì)在今年選擇Geoff作為獲獎(jiǎng)人。

英文原件內(nèi)容

Dear ACM Turing Award Committee Members:

I am writing to give my strongest recommendation to support Geoff Hinton’s nomination for Turing Award. This is the decade of Artificial Intelligence, and there is no one more qualified than Geoff in AI.

I am the Chairman and CEO of Sinovation Ventures, and have previously held executive positions at Apple, Microsoft, SGI, and Google. I was an assistant professor at Carnegie Mellon, and also received my Ph.D. there. I got to know Geoff at Carnegie Mellon, when I entered as a Ph.D. student in 1983. I took classes on neural networks from him, worked with his research team on applying neural networks to speech recognition, and was supervised by him for my minor thesis (on applying Bayesian learning to game-playing), and consulted him and his team on my Ph.D. thesis (machine learning approach to speech recognition).

While Geoff was not my Ph.D. advisor, his impact on my Ph.D. thesis was tremendous. His student Peter Brown (co-inventor of statistical machine translation, now CEO of Renaissance Technologies) was my mentor, and taught me how to apply various types of machine learning algorithms to speech recognition. This was a primary reason that helped my Ph.D. thesis to become the best-performing speech recognizer in 1988, which helped shift the speech recognition field from expert-systems approach to machine-learning approach. If I have benefited so much from Geoff and Peter, there must be thousands of other beneficiaries, given Geoff’s brilliance, persistence, and generosity.

Geoff’s contributions to AI are immense and incomparable. In my recent best-selling book AI Supowerpowers: China, Silicon Valley, and the New World Order, I described Geoff’s contribution as follows, in layman’s language:

Deep learning’s big technical break finally arrived in the mid-2000s, when leading researcher Geoffrey Hinton discovered a way to efficiently train those new layers in neural networks. The result was like giving steroids to the old neural networks, multiplying their power to perform tasks such as speech and object recognition.

Soon, these juiced-up neural networks—now rebranded as “deep learning”—could outperform older models at a variety of tasks. But years of ingrained prejudice against the neural networks approach led many AI researchers to overlook this “fringe” group that claimed outstanding results. The turning point came in 2012, when a neural network built by Hinton’s team demolished the competition in an international computer vision contest.

In the twelve years since Geoffrey Hinton and his colleagues’ landmark paper on deep learning, I haven’t seen anything that represents a similar sea change in machine intelligence.

In terms of academic accomplishments, Geoff has more than 250,000 citations, with more than half in the last five years. He has an astoundingly high H-index of 142. He was the co-inventor of the seminal work on Boltzmann Machines and backpropagation using gradient descent (published in 1983 with Terry Sejnowski, and the Nature paper with David Rumelhart in 1986). This work introduced the idea of hidden layers in neural networks, along with a mathematically elegant and computational tractable way to train their affiliated parameters. Hidden layers freed the software from “human control” (such as expert systems) and back propagation allowed non-linear combination to essentially discover prominent features (in a more goal-directed way than humans) in the process. However, it turned out that these ideas were before their time, as there were not enough data or computing power to enable these theoretical approaches to solve real-world problems or beat other approaches in competitions. The early-1980’s were dominated by expert systems, which became discredited in by late-1980’s when they were proven to be brittle and unscalable. What displaced expert systems was not Geoff’s proposals (which were too early), but simplified versions of neural networks which were compromised to work with less data and computation. My Ph.D. thesis (using hidden Markov models) was among them, and these simplified approaches were able to make some contributions with some applications, but like expert systems, they were not able to scale to the hardest problems (such as playing Go, human-level speech or vision).

From 1985 to 2015, the amount of data and computation increased tremendously. For example, my 1988 Ph.D. thesis used the largest speech database at the time, which was only 100 MB. Today, the best speech recognition systems are trained on 100 TB of data – a million-fold increase. And with that much increase in data size, Geoff’s approach (later re-branded deep learning) finally shined, as it could increase the number of layers from one to thousands, and deep learning systems continued to improve as the data size and the complexity of the model increased.

This is easy to see in hind sight. But at the time, the reality was quite cruel. The 1990s were the darkest hours for neural network researchers like Geoff. Geoff’s earlier theoretical work created intellectual spark, but the lack of data and computation prevented these deep learning systems from demonstrating superior performance. Funding dried up, and many neural network researchers moved away to other areas. But Geoff was among the few researchers (other key researchers include Yann LeCun and Yoshua Bengio) who persisted on pushing forward the neural network approach, despite a frosty and fickle environment for funding. He moved to Canada, adjusted his group to a smaller funding environment, and continued to push the frontier.

His contribution to the neural network approach continued in the next 30 years, including the mixture of experts model, the Helmholtz machine, the neural-animator system, probabilistic inference, stochastic neighbor embedding and neighborhood component analysis, t-SNE, and many other innovative ideas.

Very few technologies have disrupted multiple areas of research completely, but deep learning did. From 2010 to 2016, essentially the entire field of perception – speech recognition, image recognition, computer vision, switched to deep learning, as Geoff and his colleagues proved deep learning to be the best and most generalizable approach for perception. In the entire fiend of Artificial Intelligence, human perception (to hear, see, and understand) was considered one of the aspects that set the humans apart and a grand challenge for AI (incidentally, playing Go was another, which was conquered by Deepmind’s AlphaGo, which also used deep learning during the matches which shocked the world, and was another catalyst for the “AI revolution”).

Here is how Geoff’s team disrupted computer vision research, In 2012, his team built a neural-network based system that cut the error rate by 40% on ImageNet’s 1000-object recognition task and competition. The computer vision community was accustomed to incremental improvements annually. Geoff’s team’s results shocked that community, as a relative “outsider” using an “unconventional approach” won by such a large margin. If backpropagation was Geoff’s most important theoretical contribution, his team’s work on the ImageNet competition was Geoff’s most recognized contribution. That ImageNet result started the first ripple that ultimately became the deep learning tidal wave.

The deep learning tidal wave (the most central part of the “AI revolution”) is now changing every industry. As an example, as a venture capitalist in China, I was a part of a “tiny” side effect: Geoff’s 2012 paper and ImageNet result inspired four computer vision companies in China, and today they are collectively worth about $12 billion. Keep in mind, this was just one small field in one country based on one of Geoff’s result. Geoff’s result also led to deep learning disrupting speech recognition (the area of my Ph.D. work), resulting in super-human accuracy in 2015 by Baidu’s Andrew Ng (recruited to Baidu after Geoff joined Google part-time). And much more broadly, every technology monolith (Google, Microsoft, IBM, Facebook, Amazon, Baidu, Tencent, Alibaba) built its platform for deep learning, and re-branding themselves as “AI companies”. And in venture capital, we saw the emergence of many unicorns (in China alone there are over twenty) powered by deep learning. Also, deep learning required much compute power that traditional CPUs could not handle, which led to the use of GPUs, the rise of Nvidia and the re-emergence of semiconductors to handle deep learning work-load. Most importantly, our lives have changed profoundly – from search engines to social networks to e-commerce, from autonomous stores to autonomous vehicles, from finance to healthcare, almost every imaginable domain is either being re-invented or disrupted by the power of machine learning. In any domain with sufficient data, deep learning has led to large improvements in user satisfaction, user retention, revenue, and profit. The central idea behind deep learning (and originally from backpropagation) that an objective function could be used to maximize business metrics has had profound impact on all businesses, and helped the companies that have data and embraced machine learning to become incredibly profitable.

In aggregate, Artificial Intelligence (AI) is arguably the most exciting technology ripe for applications today. PWC and McKinsey predicted that AI would add $12-16 trillion to the global GDP by 2030. The most important advance and the primary reason that AI is believed to have matured is Geoff’s work on deep learning. While every Turing Award recipient has made seminal impact to Computer Science, few have changed the world as Geoff is doing.

Beyond the role of an innovator, Geoff was also a true thought leader. While he is soft-spoken, he is a spiritual leader who really shapes and reshapes the overall research community. He was a tireless in teaching not only his students but the world. For example, he started the Connectionist School in 1986. He personally connected to and persuaded people in computer vision and speech processing to understand and embrace deep learning. Yet, after all that work succeeded, and the world was won over by deep learning in 2018, he set a new direction. Because industry has rallied around deep learning, and large companies were gathering more data and leading the “industrialization” of deep learning, he made an exhortation to move on and focus on inventing “the next deep learning”, or fundamentally new approach to AI problems that could move closer to true human intelligence.

His thought leadership was grounded in his life-long vision and quest to better understand human cognition. While deep learning is a breakthrough that is changing the world, he sees it as only a stepping stone towards the realization of his long-term vision. To set another example, his new work on capsule leaning is again causing researchers to rethink their role and responsibilities in Geoff’s vision.

I believe Geoff is the single most important figure in the field of Artificial Intelligence today. His contributions to academia and industry are equally outstanding. He is not only a brilliant and inspirational scholar, but also an inquisitive, generous, persistent, decent, and principled gentleman, who is a role model for any aspiring young computer scientist. His work went well beyond neural networks and machine learning, and has greatly impacted computer vision, speech and signal processing, statistics, cognitive science, and neural science. I cannot think of anyone else more deserving of the Turing Award, and urge the committee to select Geoff as the recipient this year.

Sincerely,

Kai-Fu Lee, Ph.D.

Chairman & CEO, Sinovation Ventures

Fellow, IEEE

Honorary Ph.D., Carnegie Mellon University

Honorary Ph.D., City University of Hong Kong

— 完 —

誠(chéng)摯招聘

量子位正在招募編輯/記者，工作地點(diǎn)在北京中關(guān)村。期待有才氣、有熱情的同學(xué)加入我們！相關(guān)細(xì)節(jié)，請(qǐng)?jiān)诹孔游还娞?hào)(QbitAI)對(duì)話界面，回復(fù)“招聘”兩個(gè)字。

量子位 QbitAI · 頭條號(hào)簽約作者

?'?' ? 追蹤AI技術(shù)和產(chǎn)品新動(dòng)態(tài)