Many Flavors Exist in Quantum Computing Designs

Quantum physics, especially applied to computing, offers great promise for the military, national security and the commercial industry. At this point in the trajectory of the so-called Quantum 2.0 revolution, however, cost, scalability, size, power and technical considerations abound, said quantum experts speaking March 12 on a panel at AFCEA International’s inaugural TechNet Emergence conference held in Reston, Virginia, March 11-12.

The panel included John Burke, principal director for quantum science, Office of the Undersecretary of Defense for Research and Engineering, U.S. Department of Defense (DoD); Scott Buchholz, chief technology officer, global quantum computing leader, Deloitte; Joseph Broz, vice president for quantum growth and market development, IBM; Jeremy Levy, distinguished professor of physics at the University of Pittsburgh and co-founder, Pitt Quantum Institute; and was moderated by Al Mink from the AFCEA Technology Committee.

Compared to traditional computing, with a central processing unit, motherboard, memory, etc., quantum computers will be different in their functioning and certainly in size and price. And while traditional computing solves things mathematically, in a binary basis of 1s and 0s, quantum computing relies on the properties of physics to perform tasks. It is not some supercomputer running through all the possibilities. It is working differently to figure out the answers to questions, Buchholz clarified.

“Sometimes people's mental model of quantum computers is that they're going to be super supercomputers, like what we have today, only more so,” he offered. “And what I find is that, in part largely, because that's actually not true, people's intuition about what quantum computers might be used for is often misleading . . . The answer is it's working differently because it's using physics as opposed to math to figure [things] out. And what that means is, they are going to be able to solve some problems that are not tractable to math.”

The problem or challenge, however one sees it, is that the information technology industry has spent the last many decades preparing applications and solutions for traditional computers. This infrastructure does not exist for quantum computing, nor do the methods of how to apply it.

“We are still trying to figure out what problems we can map to physics,” Buchholz acknowledged. “Because we have seven or eight decades of the smartest people in the world mapping problems to math, we have gotten really good at that. It's actually difficult and we're just at the very beginning. And we have only a handful of people globally trying to figure out what problems to map to physics. That’s a lot of what makes this really interesting.”

“It still is a long way to go,” Burke echoed. “It turns out it is one of these ‘devil in the details’ kinds of situations, and you might believe me that when it comes to quantum, there are many devils and many details to be sorted out.”

The experts explained that companies are taking widely different approaches to building quantum computers, with Buchholz walking through some examples. With the industry growing so rapidly and entrants to the field joining almost every year, the list was meant only as a subset of the industry, according to Deloitte.

Moreover, the panel emphasized that quantum physics can be applied to uses other than computing, such as in sensing and communications, such as applications in timing, clocks and antennas. Within quantum computing, capabilities could be applied to optimization and post-quantum encryption, among other use cases. Any quantum computing applications, however, would have to be for gains above what traditional computing offers, the experts said.

The power of a quantum computer is measured in quantum bits, or qubits, with some current prototypes of 100 qubits and lofty industry goals of reaching 10,000 or 1 million qubits in the next few years.

IBM Quantum is pursuing perhaps the best-known method—superconducting of electrical circuits—as are Google, AWS and Rigetti. Another form, called annealer, pursued by D-Wave One, Fujitsu or ParityQC, involves the process of quantum annealing, or using optimization to find an optimal solution across a large number of solutions, according to AIMultiple research.

Photonics, or the use of programmable optical circuits, is being leveraged by companies such as Xanadu, ORCA or PsiQuantum. Palo Alto-based PsiQuantum claims that the photonics-based architecture for its quantum computers allows for manufacturing in a conventional silicon chip factory.

Intel, Diraq and Quantum Motion are relying on so-called electron spin-based silicon quantum dots. Sydney, Australia-based Diraq is claiming its technology will achieve 1 billion qubits of quantum computing.

Meanwhile, companies such as Quantinuum, Alpine Quantum Technologies (AQT) and IonQ are leveraging a trapped ion approach to quantum computing, where the technology uses electric fields and charged particles to trap single-charged atoms (or ions) inside vacuum chambers. Each ion represents a qubit, and the qubits are manipulated and measured by precisely timed laser pulses, according to documents about AQT’s PINE System.

In addition, the French company PASQAL is building neutral atom-based quantum processors from ordered atoms in 2D and 3D arrays. QuEra Computing Inc., with its 256-qubit machine and Atom Computing have similar neutral-atom approaches. In a March 12 report, PASQAL said that it is already delivering quantum computers with more than 100 qubits to its clients and hopes to scale the neutral atom quantum computers to 10,000 qubits in 2026.

The panel sees annealer, superconducting and trapped ion technologies as more mature than photonics and electron-spin silicon quantum dot computing.

The approaches, however, hinge on effective quantum error correction (QEC). Errors happen naturally in quantum computing given the physics, environmental interference, or noise, and hardware deficiencies. Such errors have to be corrected in quantum computing for the emerging technology to be more effective. Recent research has demonstrated promising paths to QEC, especially for neutral atom-related solutions.

“More and more people are finding ways to quiet down these errors through a process called quantum error mitigation,” Broz explained. “And that is leading to some practical and useful calculations that these machines even today can perform. The path that we're taking at IBM Quantum, which is called error mitigation, is using software techniques to mitigate those errors, much like noise-canceling headphones can cancel out ambient noise. That is the way that we can handle the noise or error in these quantum computers and, therefore, prolong the quantum states and derive useful measurements out of the machines that we currently have.”

In addition, quantum computers have to prove their usefulness in terms of cost as well as performance when compared to traditional computing, the experts said. Potential end users, such as those in the Department of Defense, have not seen many cost estimates.

“It actually has to deliver not just performance advantage, but it has to be economical, and I think there's still a lot of question marks about what the economics of these are going to be. These are not simple machines,” Burke said.

IBM Quantum is confident that they are reaching this performance measure, Broz noted. “We now have strong evidence that we are performing calculations that, in fact, are better than classical machines and that we are using fewer computational resources,” he said.

The industry as a whole, however, will have to grow in scalability and resiliency for the DoD to be a willing customer.

“We are going to buy a quantum computer someday for some particular application or some set of applications and we need to be pretty confident that it's going to deliver some utility, and we are very far from that confidence level,” Burke stated. “The good news is that thanks to IBM, and a number of programs, such as at the Defense Advanced Research Projects Agency and from academia all over the world, really, the case is getting better and better all the time.”