Quantum computing has a hype problem | MIT Technology Review

The most advanced quantum computers today have dozens of decorating (or “noisy”) physical quits. Building a quantum computer that can crack RSA code from such components would require billions, if not millions, of cubits. Of these, only ten thousand will be used for calculations કહે so-called logical cubits; The rest will be required for error correction, compensation for incontinence.

The Cubit systems we have today is a tremendous scientific achievement, but it brings us closer to a quantum computer that can solve any problem that anyone cares about. It’s like trying to make today’s best smartphone using a vacuum tube since the early 1900’s. You can put 100 tubes together and establish the principle that if you can somehow get 10 billion of them to work together in a consistent, seamless way, you can achieve all kinds of miracles. What is missing, however, is the success of the integrated circuit and CPU that led to the smartphone પ્રક્રિયા the discovery of a transistor that did not involve any new physics in the process took 60 years of very difficult engineering time to get to the smartphone.

There are ideas in fact, and in an approach called topological quantum computing, to bypass quantum error correction using more-stable qubits, I played some role in developing theories for these ideas. Microsoft is working on this approach. But it turns out that developing topological quantum-computing hardware is also a big challenge. It is unknown at this time what he will do after leaving the post.

Physicists are as smart as we all know (ad: I’m a physicist), and even some physicists are very good at coming up with accurate-sounding acronyms. The great difficulty in getting rid of decoherence has led to the impressive acronym NISQ for “Noisy Intermediate Scale Quantum” computers – for the idea that a small collection of noisy physical qubits could do something more useful and better than a classical computer. Not sure what this substance is: how much noise? How many cubits? Why this computer? What appropriate problems can such a NISQ machine solve?

A recent laboratory experiment at Google observed some hypothetical aspects of quantum dynamics (known as “time crystals”) using 20 noisy superconducting quits. The experiment was an impressive demonstration of the electronic control technique, but it showed no computing advantage over a conventional computer that could easily simulate time crystals with the same number of virtual quits. It also reveals nothing about the basic physics of crystals of the time. Other NISQ victories are recent experiments mimicking random quantum circuits, again a highly specialized function without any commercial value.

The use of NISQ is definitely an excellent new basic research idea – it can help in physics research in basic fields like quantum dynamics. But despite the constant drumbeat of NISQ hype coming from various quantum computing startups, the potential for commercialization is not clear. I have seen vague claims about how NISQ can be used for rapid optimization or even for AI training. I’m not an expert in optimization or AI, but I’ve asked the experts, and they’re just as mysterious. I have asked researchers at various startups how NISQ would optimize any difficult task associated with real-world applications, and I interpret their confusing answers basically by saying that we understand classical machine learning and how AI really works. Can’t, it’s possible. NISQ can do this much faster. Maybe, but this is the hope of the best, not of technology.

There are proposals to use small-scale quantum computers to form medicine, as a way to quickly calculate molecular structure, which is a surprising application because quantum chemistry is a small part of the whole process. There are equally confusing claims that quantum computers will help finance in the near term. None of the technical papers show for sure that small quantum computers, let alone NISQ machines, could lead to significant optimization in algorithmic trading or risk assessment or arbitrage or hedging or targeting and forecasting or asset trading or risk profiling. This, however, did not stop many investment banks from jumping on the quantum-computing bandwagon.

Real quantum computers today will have incredible applications, such as when the first transistor was made in 1947, no one could have predicted how it would eventually lead to smartphones and laptops. I am for hope and have great faith in quantum computing as a potentially disruptive technology, but it is very confusing to me to claim that it will in the near future start generating millions of dollars in profits for real companies selling services or products. how?

Quantum computing is indeed the most important development not only in physics, but in the whole of science. But “engagement” and “superposition” are not magic wands that we can shake and anticipate changes in technology in the near future. Quantum mechanics is really weird and contradictory, but it does not guarantee revenue and profit in itself.

A decade and a half ago, I was repeatedly asked when I thought a real quantum computer would be created. (It’s interesting that I don’t have to deal with this question anymore because the quantum-computing hype has obviously convinced people that these systems already exist or are just around the corner). My vague answer was always that I don’t know. It is impossible to predict the future of technology – it happens when it does. One can try to draw parallels with the past. It took the aviation industry more than 60 years to reach the jumbo jet, carrying hundreds of passengers thousands of miles from the Wright brothers. The immediate question is where quantum computing development, as it is today, should be placed on the timeline. Is it with the Wright brothers in 1903? The first jets around 1940? Or maybe we’re still back in the early 16th century, with Leonardo da Vinci’s flying machine? I do not know Neither anyone else.

Shankardas is the director of Sarma Condensed Matter Theory Center University of Maryland, at College Park.

Similar Posts

Leave a Reply

Your email address will not be published.