譯科技 | 量子計(jì)算或在2023年實(shí)現(xiàn)跨越式發(fā)展
來源：VentureBeat 時(shí)間：2023-03-01 12:55:51 作者：Richard Murray
近日，谷歌宣布在量子計(jì)算機(jī)糾錯(cuò)技術(shù)方面取得重要突破，引起廣泛的關(guān)注和熱議。作為一項(xiàng)新興技術(shù)，量子計(jì)算一直被認(rèn)為是量子科技領(lǐng)域的重要研究方向之一。量子計(jì)算不僅可以增強(qiáng)計(jì)算機(jī)功能，更有可能重塑應(yīng)對(duì)氣候變化、饑餓和疾病的能力，破解和更新常見的加密技術(shù)，帶來巨大經(jīng)濟(jì)效益，對(duì)全球數(shù)字經(jīng)濟(jì)產(chǎn)生影響，并形成地緣政治戰(zhàn)略意義。因此，量子計(jì)算也被全球主要經(jīng)濟(jì)體視為一項(xiàng)戰(zhàn)略技術(shù)。
ORCA computing聯(lián)合創(chuàng)始人兼首席執(zhí)行官，UKQuantum董事長兼董事Richard Murray近期發(fā)表評(píng)論文章表示，2023年或是量子計(jì)算突破式發(fā)展的一年。
2023 could be the breakthrough year for quantum computing
2022 has been a dynamic year for quantum computing. With commercial breakthroughs such as the UK Ministry of Defence (MoD) investing in its first quantum computer, the launch of the world’s first quantum computer capable of advantage over the cloud and the Nobel Prize in Physics awarded for ground-breaking experiments with entangled photons, the industry is making progress.
2022年對(duì)于量子計(jì)算來說是充滿活力的一年。隨著商業(yè)的不斷突破，如：英國國防部（MoD）投資其第一臺(tái)量子計(jì)算機(jī)，推出世界上第一臺(tái)能夠在云端發(fā)揮優(yōu)勢的量子計(jì)算機(jī)，以及因糾纏光子突破性實(shí)驗(yàn)而獲得諾貝爾物理學(xué)獎(jiǎng)……量子計(jì)算行業(yè)正取得突破性進(jìn)展。
At the same time, 2022 saw the tremendous accomplishment of the exaflop barrier broken with the Frontier supercomputer. At a cost of roughly $600 million and requiring more than 20 megawatts of power, we are approaching the limits of what classical computing approaches can do on their own. Often for practical business reasons, many companies are not able to fully exploit the increasing amount of data available to them. This hampers digital transformation across areas most reliant on high-performance computing (HPC): healthcare, defense, energy and finance.
與此同時(shí)，2022年Frontier超級(jí)計(jì)算機(jī)打破了一百萬萬億次(exaflop)超算屏障，獲得了巨大成就，這樣的成本大約為6億美元，以及超過20兆瓦的電力，我們正在接近經(jīng)典計(jì)算方法自身能夠做到的極限。通常出于實(shí)際的商業(yè)原因，許多公司無法充分利用他們所掌握的越來越龐大的數(shù)據(jù)。這大大阻礙了醫(yī)療、國防、能源和金融等最依賴高性能計(jì)算（HPC）的領(lǐng)域?qū)崿F(xiàn)數(shù)字化轉(zhuǎn)型。
To stay ahead of the curve, 91% of global business leaders are investing or planning to invest in quantum computing. According to reports, 70% are developing real-life use cases and 61% are planning to spend $1 million or more over the next three years.
為了保持領(lǐng)先地位，91%的全球商業(yè)領(lǐng)袖正在投資或計(jì)劃投資量子計(jì)算。有關(guān)報(bào)告顯示，70%的人正在開發(fā)現(xiàn)實(shí)量子計(jì)算在生活中的應(yīng)用案例，61%的人計(jì)劃在未來三年內(nèi)花費(fèi)100萬美元或更多在量子計(jì)算的研發(fā)上。
As the technology becomes more exciting and the industry gathers pace, the pressure is on for quantum to deliver. But the voice of skeptics will also grow louder. In the face of those that say quantum computers will never be useful due to their complexity and limited results to date, the question on everyone’s mind is, will 2023 be a breakthrough year for quantum computing?
隨著技術(shù)的演變，行業(yè)也以激動(dòng)人心的步伐加速發(fā)展，量子化的壓力隨之而來。懷疑者的聲音也越來越多，面對(duì)那些說量子計(jì)算機(jī)由于其復(fù)雜性和迄今為止有限成果以至于它永遠(yuǎn)派不上用場的人，每個(gè)人心中的疑問是：2023年是否能成為量子計(jì)算實(shí)現(xiàn)突破的一年？
Technical innovations vs market incumbents
技術(shù)創(chuàng)新與市場占有者
During 2022, we saw the creation of many industry incumbents who used SPACs, IPOs, mergers or corporate sponsorship to build themselves substantial war chests to pursue some serious engineering activity. While these significant scale-up activities will continue, 2023 will also be the year of innovation and possible disruption.
2022年期間，我們看到許多行業(yè)老牌企業(yè)的誕生，他們利用特殊目的收購公司、首次公開募股、合并或企業(yè)贊助來為自己建立大量資金，以從事一些嚴(yán)肅的工程活動(dòng)。雖然這些重要的擴(kuò)大活動(dòng)將繼續(xù)進(jìn)行，但 2023 年也將是創(chuàng)新和可能顛覆的一年。
Amongst the big players, new players will emerge with alternative approaches towards quantum computing: Perhaps replacing qubits and gate models with qumodes, using model simulations and quantum annealing models.
在這些大玩家中，新公司將以替代性的方法出現(xiàn)在量子計(jì)算中：也許用量子密碼代替量子比特和門模型，使用模型模擬和量子退火模型。
The aim of these newcomers will not be to solely achieve universal computing, but rather more specific and useful computation that can be delivered in a shorter timescale. The challenge will be whether these new machines can be applied to something useful and that the industry will care about them in the near term. The quantum supply chain is also developing with component-based suppliers — such as quantum processor vendors — that will shake loose how full-stack systems are built and break the economics of current black box approaches.
這些初來乍到者的目標(biāo)不僅是實(shí)現(xiàn)通用計(jì)算，而是更具體和有用的計(jì)算，可以在更短時(shí)間內(nèi)交付。挑戰(zhàn)在于這些新機(jī)器是否能應(yīng)用于有用的東西，以及行業(yè)在短期內(nèi)是否會(huì)關(guān)心它們。量子供應(yīng)鏈也在與基于組件的供應(yīng)商（例如量子處理器供應(yīng)商）一起發(fā)展，這將擺脫全棧系統(tǒng)的構(gòu)建方式，并打破當(dāng)前黑盒方法的經(jīng)濟(jì)性。
Such work will force further discussions about the right and best way to compare and benchmark technologies, performance and the industry.
此類工作將推動(dòng)進(jìn)一步討論比較和基準(zhǔn)化技術(shù)、性能和行業(yè)的正確和最佳方式。
Competition for financing
融資競爭
Despite the turmoil in the international financial markets, quantum computing may continue to buck the trend with large funding rounds. Also, 2023 will see an interesting comparison between public and privately owned quantum companies.
盡管國際金融市場動(dòng)蕩不安，但量子計(jì)算可能會(huì)繼續(xù)逆勢而上，進(jìn)行大輪融資。另外，2023年，將看到公有和私有量子公司之間有趣的較量。
Public companies will continue to put their capital to work, but at the cost of the short-term attention of investors and short sellers. While they and the rest of the industry push to meet meaningful and substantial technical milestones, they will have only partial success in shrugging off the short-term pressures to validate the business. It’s likely that a race to capture the first market share to meet revenue predictions will ensue.
上市公司將繼續(xù)把他們的資本用于工作，但代價(jià)是投資者和賣空者的短期關(guān)注。雖然他們和業(yè)內(nèi)其他人努力實(shí)現(xiàn)有意義和實(shí)質(zhì)性的技術(shù)里程碑，但他們?cè)跀[脫短期壓力以驗(yàn)證業(yè)務(wù)方面只會(huì)取得部分成功。很可能隨之而來的是爭奪第一個(gè)市場份額以實(shí)現(xiàn)收入預(yù)測的競賽。
In the private space, and with a global recession looming, large companies’ valuations will likely struggle to compete with previous expectations. This will be countered to an extent by the increasing appetite for deep tech, as well as new, exciting developments. Within the recent glut of new quantum companies, many will struggle, and both successful and less successful companies will be acquired as the big players consolidate. In general, 2023 will likely end with fewer quantum companies than in 2022.
在私營領(lǐng)域，全球經(jīng)濟(jì)衰退迫在眉睫，大型公司的估值可能將難以與之前的預(yù)期競爭。這將在一定程度上被對(duì)深度技術(shù)以及新的、令人興奮的發(fā)展的日益增長的欲望所抵制。在最近大量的新量子公司中，許多公司將陷入困境，隨著大公司的整合，成功和不太成功的公司都將被收購?？偟膩碚f，2023年結(jié)束時(shí)，量子公司的數(shù)量可能會(huì)比2022年少。
For both public and private quantum companies, it will help when a few make strides toward creating useful cases with near-term quantum computers. In the pursuit of pragmatic value-creation, this will come in many forms — including quantum sensing and comms, quantum-inspired, and hybrid quantum-classical approaches with small-scale systems.
對(duì)于公共和私人量子公司而言，當(dāng)少數(shù)公司在使用近期量子計(jì)算機(jī)創(chuàng)建有用案例方面取得進(jìn)展時(shí)，這將有所幫助。在追求務(wù)實(shí)的價(jià)值創(chuàng)造過程中，這將以多種形式出現(xiàn)——包括量子傳感和通信、量子啟發(fā)以及具有小規(guī)模系統(tǒng)的混合量子經(jīng)典方法。
A few successes here will be industry-changing, which will start to bring about a focus that the industry has been waiting for. The consequences will ripple through the entire market.
這些成就將改變行業(yè)，成為行業(yè)一直等待的轉(zhuǎn)折點(diǎn)，而后波及整個(gè)市場。
Making progress toward fault-tolerant machines
在容錯(cuò)機(jī)器方面取得進(jìn)展
Despite progress on short-term applications, 2023 will not see error correction disappear. Far from it, the holy grail of quantum computing will continue to be building a machine capable of fault tolerance. 2023 may create software or hardware breakthroughs that will show how we’re closer than we think, but otherwise, this will continue to be something that is achieved far beyond 2023.
盡管在短期應(yīng)用方面量子計(jì)算取得了進(jìn)展，但2023年將糾錯(cuò)不會(huì)消失。遠(yuǎn)非如此，量子計(jì)算的藍(lán)圖將繼續(xù)是建立一個(gè)能夠容錯(cuò)的機(jī)器。2023年可能會(huì)創(chuàng)造出軟件或硬件的突破，這比我們想象的更接近，除此之外，要遠(yuǎn)在2023年之后才能實(shí)現(xiàn)。
Even though it’s everything to some quantum companies and investors, the future corporate users of quantum computing will largely see it as too far off the time horizon to care much. The exception will be government and anyone else with a significant, long-term interest in cryptography.
盡管這對(duì)一些量子公司和投資者來說意味著一切，但量子計(jì)算未來的企業(yè)用戶將在很大程度上認(rèn)為它離時(shí)間跨度太遠(yuǎn)而不太關(guān)心。政府和其他對(duì)密碼學(xué)有重大、長期興趣的人將是例外。
However, regardless of those long time horizons, 2023 will define clearer blueprints and timelines for building fault-tolerant quantum computers for the future. Indeed, there is also an outside chance that next year will be the year when quantum rules out the possibility of short-term applications for good, and doubles down on the 7- to 10-year journey towards large-scale fault-tolerant systems.
然而，不管這些漫長的時(shí)間跨度如何，2023年將為未來構(gòu)建容錯(cuò)量子計(jì)算機(jī)確定更清晰的藍(lán)圖和時(shí)間表。事實(shí)上，明年也有可能是量子永遠(yuǎn)排除短期應(yīng)用可能性的一年，并在 7 到 10 年的大規(guī)模容錯(cuò)系統(tǒng)之旅中加倍努力。
Governments, users and HPC
政府、用戶和手持電腦
2022 saw the German government conclude the tendering process for some very large quantum computing projects, with one example of a €67m contract for two projects. In 2023, that trend will continue with yet more public procurements for quantum computing.
2022 年，德國政府完成了一些超大型量子計(jì)算項(xiàng)目的招標(biāo)過程，其中一個(gè)例子是兩個(gè)項(xiàng)目的 6700 萬歐元合同。到 2023 年，隨著更多的量子計(jì)算公共采購，將有更多的量子計(jì)算公共采購。
Those tenders and the fact that they will be run through several of the world’s HPC centers will force the quantum computing industry to live up to the rigor of tender requirements, and the delivery obligations which come with it. So long as those tenders are run well, these activities will force up the maturity of the technology, and the companies in this space.
這些招標(biāo)以及它們將通過全球多個(gè) HPC 中心進(jìn)行的事實(shí)，將迫使量子計(jì)算行業(yè)達(dá)到嚴(yán)格的招標(biāo)要求，以及隨之而來的交付義務(wù)。只要這些招標(biāo)運(yùn)行良好，這些活動(dòng)將迫使該技術(shù)和該領(lǐng)域的公司變得更加成熟。
Alongside that, the sophistication of the user community will develop dramatically this year. Expect the launch of several ‘industrial challenges’ delivered by teams of in-house quantum experts. Again, this increasing maturity will act as a force for good within the industry, helping to create great strides toward the search for concrete applications and roadmaps.
除此之外，今年用戶社區(qū)的成熟度將顯著提高。期待由內(nèi)部量子專家團(tuán)隊(duì)發(fā)起的多項(xiàng)“工業(yè)挑戰(zhàn)”。同樣，這種日益成熟的技術(shù)將成為行業(yè)內(nèi)的一股良性力量，有助于在尋找具體應(yīng)用和路線圖方面取得長足進(jìn)步。
Geopolitics standing in the way
地緣政治阻礙
Geopolitics will continue to shape quantum as it does the rest of the economy; this shaping could reach a fever pitch with the growing separation between the U.S. and China. As the race is on to develop quantum computers to gain a strategic lead in cybersecurity, intelligence operations and the economic industry, expect increasing restrictions to limit technological exchange and increasing impact on supply chains. This will be partially offset through bi and multilateral agreements between nations, although the specter of nationalism will linger.
地緣政治將繼續(xù)影響量子，就像它影響其他經(jīng)濟(jì)領(lǐng)域一樣。隨著大國之間的較量日益嚴(yán)峻，這種阻礙可能會(huì)達(dá)到白熱化的程度。隨著開發(fā)量子計(jì)算機(jī)以在網(wǎng)絡(luò)安全、情報(bào)行動(dòng)和經(jīng)濟(jì)產(chǎn)業(yè)中獲得戰(zhàn)略領(lǐng)先地位的競賽正在進(jìn)行，預(yù)計(jì)越來越多的限制將限制技術(shù)交流和對(duì)供應(yīng)鏈的影響越來越大。這將通過國家間的雙邊和多邊協(xié)議部分抵消，盡管民族主義的幽靈將揮之不去。
But how will European and UK companies fare? Many are fearful of being caught up in the middle of the China-U.S. tech competition, and so are urgently designing quantum tools to protect their interests.
歐洲和英國的公司將如何應(yīng)對(duì)？許多人擔(dān)心被卷入大國科技競爭的漩渦中，因此正在緊急設(shè)計(jì)量子工具以保護(hù)自己的利益。
A breakthrough year for quantum
量子突破的一年
So as we look forward, it’s no longer a question of if quantum computing will be available but when? 2023 may be the year in which some ask — and perhaps even claim now. Whereas others will continue to say, ‘of course not.’’
因此，當(dāng)我們展望未來，這不再是一個(gè)是否會(huì)有量子計(jì)算的問題，而是何時(shí)可用？2023 年也許是一些人提出疑問或者聲稱能夠?qū)崿F(xiàn)的一年，而其他人則會(huì)繼續(xù)說，“當(dāng)然不是?！?/p>
因此，當(dāng)我們向前看時(shí)，這不再是一個(gè)是否會(huì)有量子計(jì)算的問題，而是什么時(shí)候？2023年可能是一些人問的年份--甚至可能現(xiàn)在就聲稱。而其他人則會(huì)繼續(xù)說，“當(dāng)然不會(huì)”。
With more and more companies adopting quantum to explore its potential, we will certainly leave 2023 more aware of the benefits and timeline. This may help companies better understand what their future could look like with quantum.
隨著越來越多的公司采用量子來探索其潛力，我們肯定會(huì)在 2023 年更加了解其好處和時(shí)間表。這可能有助于公司更好地了解他們的量子未來會(huì)是什么樣子。
Yet however little we know about what the future holds, one thing is certain: The world will be watching.
盡管我們對(duì)未來知之甚少，有一件事可以肯定——全世界都在關(guān)注量子計(jì)算。
責(zé)任編輯：張薇