WILL QUANTUM COMPUTERS REPLACE THEIR CLASSICAL COUNTERPARTS?

Classical computing isn’t going away, but quantum technology has the potential to disrupt many industries. It’s crucial to leverage the strengths of both to unlock quantum’s full potential.

Rhea Moutafis, Expert Columnist
January 23, 2021

The promises of quantum computing are plentiful: It could help develop lifesaving drugs with unprecedented speed, build better investment portfolios for finance and usher in a new era of cryptography. Does that mean quantum computing will become the standard and classical computing will become obsolete?

The short answer is no. Classical computers have unique qualities that will be hard for quantum computers to attain. The ability to store data, for example, is unique to classical computers since the memory of quantum computers only lasts a few hundred microseconds at most.

Additionally, quantum computers need to be kept at temperatures close to absolute zero, which is on the order of -270 degrees Celsius (-450 degrees Fahrenheit). Average consumers don’t have such powerful refrigerators at home, nor would it be advisable for them to do so when thinking about the corresponding energy consumption and its impact on the environment. All of these challenges suggest that quantum computers are unlikely to become a fixture of most households or businesses.

The likelier scenario is that researchers in academia and industry will access quantum computers through cloud services. Although quantum technology is still in its early stages, providers like Amazon Web Services and Microsoft Azure already offer cloud access to it.

There’s no doubt that quantum computing will transform many industries in the next decade. Classical computers will always play a role, however. As always, though, the devil is in the details: Which problems are better suited for quantum, and which for classical computers? Which industries will profit most from adopting a hybrid quantum-and-classical computing strategy?

QUANTUM WON’T REPLACE CLASSICAL COMPUTERS

Quantum computing has existed since the early 1980s. Four decades later, though, and we still don’t even have three dozen quantum devices worldwide. The first hands-on proof of the quantum advantage, i.e., that quantum computers are a lot faster than classical computers, was only demonstrated by Google in 2019.

According to a recent McKinsey report, we still might not even have more than 5,000 quantum machines by 2030. This isn’t just because it will be hard to store data for a long period in quantum computers or to operate them at room temperature, either. It turns out that quantum computing is so fundamentally different from classical computing that it will take time to develop, deploy and reap the benefits of the technology.

One example of such a fundamental difference is that quantum computers can’t give straightforward answers like classical computers do. Classical computations are quite simple: You provide an input, an algorithm processes it, and you end up with an output. Quantum computations, on the other hand, take a range of different inputs and return a range of possibilities. Instead of getting a straightforward answer, you get an estimate of how probable different answers are.

This style of computing can be very useful when dealing with complex problems in which you have many different input variables and complex algorithms. On a classical computer, such a process would usually take a very long time. Quantum computers could narrow down the range of possible input variables and solutions to a problem. After that, one can obtain a straightforward answer by testing the range of inputs that the quantum computer provided with a classical computer.

Classical computers will therefore remain useful for decades to come. Their continued relevance is not just a question of how long it’ll take for quantum computers to be developed enough to reach mainstream adoption, either. It’s also about the fuzzy nature of the solutions that quantum computing returns. As humans, we prefer straightforward answers, which can only be obtained by classical computers.

TECHNICAL LIMITATIONS OF QUANTUM COMPUTING

We should keep in mind, however, that this way of working hasn’t been extensively tested yet. With quantum technology still in its infancy after four decades of work, there are some quite tricky limitations to get around. These limitations will ensure classical computers remain relevant.

Information on quantum computers is stored and processed in units called qubits. Similar to bits in classical computers, they can have different values, like zero or one. Qubits, however, can also have mixtures of zero and one, like 30 percent zero and 70 percent one, for example. This ability makes them quite powerful: Whereas a classical computer with N bits can perform a maximum of N calculations at once, quantum computers can manage up to 2^N calculations. Therefore, if a classical processor manages 10 calculations, a quantum processor would manage 2^10, or 1,024 calculations.

The problem is that it’s extremely difficult to build quantum computers with many qubits – the current record is a Chinese machine with 76. To be fair, fledgling startups and tech giants alike have promised to build machines with thousands of qubits soon. Larger quantum machines tend to have lower connectivity, however, which means that the qubits can’t communicate with one another quite as well. This lack of connectivity lowers the system’s overall computing power.

Finally, quantum machines are quite error-prone. These computational errors are inherent to quantum systems and can’t be avoided per se. That’s why lots of capital and talent are being poured into quantum error detection, developing ways to build machines that notice their own mistakes and correct them. Although tremendous advances have been made in this area, it’s unlikely that quantum errors will ever fully disappear. Even with highly accurate quantum computers, verifying the end results with classical computers will remain necessary.

WAITING ON DISRUPTION

Add the technical limitations of quantum computing to the supercool temperatures necessary for storing hardware, and you start to understand why most companies are hesitant to invest in quantum computing as of now. In some industries, however, a quantum computer might become economically viable even if it solves a problem “only” 1,000 times faster than a classical computer. This includes sectors such as finance, pharma and cryptography.

It’s therefore plausible that companies will adopt quantum systems little by little as they bring increasing economic benefits. During this time, classical computers will remain relevant, vital even, to maintain the status quo. Few companies will invest big in these early days, meaning classical computing will still be doing the lion’s share of the work in industry.

On the flip side, big investments, however risky they seem now, are poised to be the driving force behind real breakthroughs in the adoption of quantum computing. Such disruptions are particularly notable in two sectors: drug development and cryptography. These two areas thrive on enormous computational capacities, which quantum machines could provide in unprecedented ways.

WHERE QUANTUM WILL THRIVE — AND CLASSICAL WILL HELP

Not all industries will profit from quantum computing in the same way. According to McKinsey, there are four areas where quantum computing could yield immense long-term gains. Nevertheless, classical computing will remain relevant in these areas and complement the benefits of quantum technology.

  • Drug Development: Computational simulations of drug molecules are essential since they cut costs and time, sometimes dramatically. Today, this type of simulation is only possible with relatively small molecules though. If, however, companies are interested in proteins, which often have thousands of constituents, they need to manufacture them and test their properties in real life because today’s computing resources are not sufficient to make an accurate simulation. Quantum simulations could dramatically reduce the costs of development and help bring drugs to the market faster. Nevertheless, since quantum computing always returns a range of possibilities, the optimal molecular structure of a drug will still need to be confirmed with a classical computer.
     
  • Optimization Problems: What is the most efficient placement of equipment in a factory? What is the best way to deploy vehicles to ensure an efficient transportation network? What is the best investment strategy for optimal returns in five, 10 or 30 years? These are complex problems for which the best answer isn’t always obvious. With quantum computers, one could dramatically narrow down the possibilities and then use classical computers to get straightforward answers. These problems abound in diverse sectors, from manufacturing to transportation to finance.
     
  • Quantum Artificial Intelligence: Billions of dollars are being invested in autonomous vehicles. The aim is to make vehicles so smart that they’re fit for busy roads anywhere on Earth. Although there is a lot of talent working on training AI algorithms to learn how to drive, accidents are still a problem. Quantum AI, which might be a lot faster and more powerful than current methods, may help solve this problem. The benefits might only be reaped in a decade from now, however, since quantum AI is a lot less developed today than quantum simulation or cryptography. Therefore, the majority of AI algorithms will continue to be deployed on classical computers. Although it’s too early to confidently predict right now, it is not unthinkable that most AI will be quantum in a couple of decades.
     
  • Cryptography: Today’s security protocols rely on random numbers and number factorization in heights that classical computers can compute to generate a password but rarely solve in order to crack a password. In a few years, though, quantum computers might be so powerful that they could crack any password. That’s why it’s imperative that researchers start investing into new, quantum-safe cryptography. Quantum technology is a double-edged sword, however, in the sense that it won’t only crack every password, but also be able to generate new, unhackable cryptographic keys. The space is moving fast to incorporate this new reality. Since classical computers will remain relevant, it’s imperative that quantum-safe cryptography exists for these too. This is possible, and companies have already started securing their data on classical computers this way.

PREPARING FOR A QUANTUM FUTURE

Google, IBM and a host of different startups are hoping to double their quantum computing capacities each year. This means that some companies need to follow the footsteps of Barclays, BASF, BMW, Dow and ExxonMobil, which are already active in quantum technology.

Clearly, not every sector is likely to benefit in the same way. Sectors such as pharma, finance, travel, logistics, global energy and materials may profit earlier than others. This also means that players in these fields need to gear up swiftly if they don’t want to be left behind by their competitors.

If it makes strategic sense, companies in these sectors should follow suit with front-runners like Barclays or ExxonMobil, and build a team of quantum talent in-house. Such talent is scarce already, and it’s quite unlikely that universities will be able to keep up with the expanding demand. As an alternative, they could consider directly partnering with companies that are developing quantum technologies, which may give them a competitive advantage down the road.

Of course, this doesn’t mean that these companies will stop using classical computers. Rather, quantum computing will bring them huge benefits on specific tasks, such as drug development, financial engineering and more.

In contrast to companies that will benefit from the emergence of quantum technology, others will have to invest in ways to stay safe from it. Specifically, companies with long-lived data assets should start investing in quantum-safe cryptography to protect themselves from future attacks. This doesn’t mean that they need to use quantum computers anytime soon, but they should make sure they’re safe from future quantum attacks. The concerned sectors range from aerospace engineering to pharmacological development to socioeconomic data for market research.

In short, companies in many different sectors need to gear up. Some because they’ll benefit from quantum if they adopt it, others because they’ll need to move ahead of the threat. Either way, companies that don’t embrace the power of this new quantum-classical hybrid risk being left behind.

Quantum computing isn’t going to take over the world. But it’s going to have a major impact in the next decade or two by working in full concertation with classical computers.

SOURCE: builtin.com
MAIN IMAGE SOURCE: pexels.com

More from the blog

WHY COMPUTERS WILL LIKELY NEVER PERFORM ABDUCTIVE INFERENCES

As Erik Larson points out in The Myth of Artificial Intelligence, what computers “know” must be painstakingly programmed WILLIAM A. DEMBSKIAPRIL 24, 2021 I’ve been reviewing...

How digital literacy can help close the digital divide

by Kara Brisson-Boivin, Samantha McAleese. Originally published on Policy OptionsApril 13, 2021 Everything from education and work to health care, social services, food and grocery...

AI-Powered Content Filtering Adds New Protection Powers

Recent innovations give content filtering greater impact and broader use in improving student learning and engagement overall for schools. by Erika Gimbel, edtechmagazine.comMAR. 22, 2021 Schools...