谷歌(Google)宣布推出一款名為Willow的新型量子芯片,這一創(chuàng)新成果引起了包括世界首富埃隆·馬斯克在內(nèi)的廣泛關(guān)注。
在周一的一篇博文中,谷歌量子人工智能部門的創(chuàng)始人兼負責(zé)人哈特穆特·內(nèi)文(Hartmut Neven)指出,Willow量子芯片為實現(xiàn)全面量子計算機鋪平了道路,它有望“通過加速科學(xué)探索的步伐、開發(fā)實用的應(yīng)用程序和解決社會面臨的部分最大挑戰(zhàn)來造福社會”。
Alphabet首席執(zhí)行官桑達爾·皮查伊(Sundar Pichai)在社交媒體發(fā)布Willow量子芯片的消息后,X公司首席執(zhí)行官埃隆·馬斯克以“哇”作為回應(yīng),并與皮查伊進行了一番交流,皮查伊表示:“或許有一天,我們能夠利用星艦在太空建立量子集群。”
“這很可能會成為現(xiàn)實。"馬斯克回答道。
為何Willow如此特別?
簡言之,Willow是一款量子芯片,比如谷歌開發(fā)的這款芯片。量子芯片具備執(zhí)行復(fù)雜計算的能力,而這些計算若由傳統(tǒng)計算機來完成,所需時間甚至可能超過宇宙的年齡。對于外行人來說,基準(zhǔn)測試程序是所有計算機都試圖解決的難題,旨在幫助人類了解哪臺計算機在解決特定問題時速度最快或效果最佳。
Willow量子芯片的獨特之處在于其具備從錯誤中學(xué)習(xí)的能力。傳統(tǒng)計算機在解決問題時經(jīng)常會出錯,而谷歌聲稱,Willow實際上會隨著量子比特(量子計算機中的基本計算單位)數(shù)量的增加而變得更加智能。這好比建造一座高聳入云的塔樓,而塔樓不僅不會因高度的增加而變得搖搖欲墜,反而愈發(fā)穩(wěn)固——至少,谷歌是這樣宣稱Willow的。哈特穆特表示,這是量子糾錯領(lǐng)域存在了近30年的難題,他堅稱Willow能夠在計算規(guī)模擴大的同時,使邏輯錯誤抑制實現(xiàn)指數(shù)級增長。
內(nèi)文表示,Willow具備實時糾錯能力:這好比擁有一塊橡皮擦,其清除錯誤速度之快,遠超你涂色超出邊界的速度。從本質(zhì)上講,Willow能夠確保計算機精準(zhǔn)無誤地解決計算問題,并在運行過程中消除錯誤。
他說:"這是迄今為止最具說服力的可擴展邏輯量子比特原型。它傳遞出一個強有力的信號,表明功能強大的超大型量子計算機的實現(xiàn)已成為可能?!?/p>
內(nèi)文認(rèn)為,Willow迄今為止取得的最大成就之一,是在五分鐘內(nèi)解決了一個數(shù)學(xué)難題。乍一看,這似乎不足為奇,但要考慮到谷歌曾聲稱,傳統(tǒng)計算機解決同樣的難題需要花費10^25年的時間。正是這一功能讓谷歌相信,Willow能夠助力人類發(fā)現(xiàn)新型藥物、構(gòu)思下一代電池,并最終助力地球節(jié)約能源。
值得一提的是,包括谷歌在內(nèi)的許多科技公司都曾對量子計算領(lǐng)域的所謂“突破性進展”夸大其詞。不過,谷歌、馬斯克和其他業(yè)界領(lǐng)袖似乎對Willow表現(xiàn)出了濃厚的興趣,以至于該公司決定提供開源軟件,并在Coursera平臺上開設(shè)一門新課程,旨在幫助開發(fā)人員、研究人員和工程師“開發(fā)能夠解決未來問題的算法”。(財富中文網(wǎng))
譯者:中慧言-王芳
谷歌(Google)宣布推出一款名為Willow的新型量子芯片,這一創(chuàng)新成果引起了包括世界首富埃隆·馬斯克在內(nèi)的廣泛關(guān)注。
在周一的一篇博文中,谷歌量子人工智能部門的創(chuàng)始人兼負責(zé)人哈特穆特·內(nèi)文(Hartmut Neven)指出,Willow量子芯片為實現(xiàn)全面量子計算機鋪平了道路,它有望“通過加速科學(xué)探索的步伐、開發(fā)實用的應(yīng)用程序和解決社會面臨的部分最大挑戰(zhàn)來造福社會”。
Alphabet首席執(zhí)行官桑達爾·皮查伊(Sundar Pichai)在社交媒體發(fā)布Willow量子芯片的消息后,X公司首席執(zhí)行官埃隆·馬斯克以“哇”作為回應(yīng),并與皮查伊進行了一番交流,皮查伊表示:“或許有一天,我們能夠利用星艦在太空建立量子集群?!?/p>
“這很可能會成為現(xiàn)實。"馬斯克回答道。
為何Willow如此特別?
簡言之,Willow是一款量子芯片,比如谷歌開發(fā)的這款芯片。量子芯片具備執(zhí)行復(fù)雜計算的能力,而這些計算若由傳統(tǒng)計算機來完成,所需時間甚至可能超過宇宙的年齡。對于外行人來說,基準(zhǔn)測試程序是所有計算機都試圖解決的難題,旨在幫助人類了解哪臺計算機在解決特定問題時速度最快或效果最佳。
Willow量子芯片的獨特之處在于其具備從錯誤中學(xué)習(xí)的能力。傳統(tǒng)計算機在解決問題時經(jīng)常會出錯,而谷歌聲稱,Willow實際上會隨著量子比特(量子計算機中的基本計算單位)數(shù)量的增加而變得更加智能。這好比建造一座高聳入云的塔樓,而塔樓不僅不會因高度的增加而變得搖搖欲墜,反而愈發(fā)穩(wěn)固——至少,谷歌是這樣宣稱Willow的。哈特穆特表示,這是量子糾錯領(lǐng)域存在了近30年的難題,他堅稱Willow能夠在計算規(guī)模擴大的同時,使邏輯錯誤抑制實現(xiàn)指數(shù)級增長。
內(nèi)文表示,Willow具備實時糾錯能力:這好比擁有一塊橡皮擦,其清除錯誤速度之快,遠超你涂色超出邊界的速度。從本質(zhì)上講,Willow能夠確保計算機精準(zhǔn)無誤地解決計算問題,并在運行過程中消除錯誤。
他說:"這是迄今為止最具說服力的可擴展邏輯量子比特原型。它傳遞出一個強有力的信號,表明功能強大的超大型量子計算機的實現(xiàn)已成為可能?!?/p>
內(nèi)文認(rèn)為,Willow迄今為止取得的最大成就之一,是在五分鐘內(nèi)解決了一個數(shù)學(xué)難題。乍一看,這似乎不足為奇,但要考慮到谷歌曾聲稱,傳統(tǒng)計算機解決同樣的難題需要花費10^25年的時間。正是這一功能讓谷歌相信,Willow能夠助力人類發(fā)現(xiàn)新型藥物、構(gòu)思下一代電池,并最終助力地球節(jié)約能源。
值得一提的是,包括谷歌在內(nèi)的許多科技公司都曾對量子計算領(lǐng)域的所謂“突破性進展”夸大其詞。不過,谷歌、馬斯克和其他業(yè)界領(lǐng)袖似乎對Willow表現(xiàn)出了濃厚的興趣,以至于該公司決定提供開源軟件,并在Coursera平臺上開設(shè)一門新課程,旨在幫助開發(fā)人員、研究人員和工程師“開發(fā)能夠解決未來問題的算法”。(財富中文網(wǎng))
譯者:中慧言-王芳
Google is turning heads—including that of world’s richest man, Elon Musk—with the announcement of its new quantum chip, called Willow.
In a Monday blog post, Hartmut Neven, founder and lead for Google Quantum AI, said the Willow chip paves the way for a full-scale quantum computer that can “benefit society by advancing scientific discovery, developing helpful applications, and tackling some of society’s greatest challenges.”
After Alphabet CEO Sundar Pichai announced Willow on social media, X CEO Elon Musk responded with a “wow” and engaged in a back and forth with Pichai, who said “we should do a quantum cluster in space with Starship one day :).”
“That will probably happen,” Musk replied.
Why is Willow so special?
To break Willow down in simple terms: A quantum chip, like the one developed here by Google, can perform computations that would take regular computers longer than the age of the universe to execute. Benchmarks, for those unfamiliar, are challenges that all computers try to solve to help us humans understand which one is the fastest, or best at solving a particular problem.
What sets Willow apart is its ability to learn from its mistakes. While regular computers can frequently err in problem-solving, Google claims Willow actually gets smarter as it uses more qubits, which are the units of computation in quantum computers. It’s like building an incredibly high tower, but instead of wobbling more as it grows taller, it gets even steadier—at least, that’s what Google is claiming about Willow. This has been a challenge that’s existed in quantum error correction for nearly 30 years, according to Hartmut, who insists Willow reduces errors “exponentially” as it scales up.
Neven says Willow can fix its mistakes in real-time: like having an eraser that works faster than you can color outside the lines. It essentially results in a computer that can solve computations perfectly, erasing mistakes as it goes.
“This is the most convincing prototype for a scalable logical qubit built to date,” he said. “It’s a strong sign that useful, very large quantum computers can indeed be built.”
One of the biggest achievements of Willow so far, according to Neven, is that it solved a mathematical puzzle in five minutes. At first glance, that doesn’t sound like a big deal, until you consider that Google claims it would take a regular computer 10 septillion years to solve that same puzzle. It’s this kind of functionality that leads Google to believe Willow can help humans discover new medicines, conceptualize next-generation batteries, and ultimately help our planet save energy.
For what it’s worth, many tech companies, including Google, have a history of exaggerated claims over so-called “breakthroughs” in quantum computing. But Google, Musk, and others seem to be very excited about Willow—to the extent the company has decided to make open-source software available, as well as a new Coursera course, to help developers, researchers, and engineers “create algorithms that can solve the problems of the future.”