The Near-Term Future of Quantum Computing? Analog Simulations

​Contrary to what you may have ​heard, progress in quantum computing is slow and painful. While there are papers and studies out every week describing some new technology that could have a quantum computing implication at some theoretical future point—together reinforcing the impression that we are really, truly almost there—they usually have less to do with the thing itself: a quantum computer.

Physicists have performed quantum calculations on small handfuls​ of qubits, the quantum analog of the classical bit information carrier, but the real prize remains distant.

There is a reason for this: a quantum computer is really, really hard to do. Mostly, this has to do with the fragility of quantum information (qubits), which, once disturbed, become classical junk. It's a hardware challenge. How do you protect a qubit—a single particle in a superposition of different states—when its very surroundings threaten to wipe it out? It's worse than that even, because wiping out one qubit means potentially wiping out the entire network of entangled qubits, nuking not just one unit of information but every parallel processing unit too.

The problem is worse still. While information gets lost in computing just as a matter of course, classical error correction algorithms don't work in quantum computing because attempting to compare qubits to assess whether or not a given piece of information has been compromised is just another way to disrupt that qubit.​

The physicist Ivan Deutsch doesn't seem too discouraged, however, even though his task for the past 20 years has been designing the guts of a quantum computer. In a great Quant​a Q&A he offers some hope in the form of old-school computing: quantum simulations, not quite the real deal, but immensely powerful nonetheless.

The key component is what Deutsch calls the "qudit." It's a complete atom rather than a single electron or other fundamental particle. It operates in a sort of fake or simulated quantum superposition in which it's allowed to occupy one of 16 different states, represented by energy levels. As Deutsch explains, using a full atom of 16 possibilities is a fundamentally different beast than a computer using the classical bit (on or off) or the quantum qubit (on and off). It's analog.

This presents its own problems, however:

What are the main challenges for these quantum simulators?
​Because the evolution of the analog simulation is not digitized, the software cannot correct the tiny errors that accumulate during the calculation as we could error-correct noise on a universal machine. The analog device must keep a quantum superposition intact long enough for the simulation to run its course without resorting to digital error correction. This is a particular challenge for the analog approach to quantum simulation.

Deutsch's analog interest isn't isolated in the quantum computing research community. With less resources available than labs at, say, IBM (etc.), analog quantum simulators offer short-term progress to academic researchers. "There is short-term fruit to be picked in that arena—both intellectually and in the currency of academics: publishable papers," he said.

Meanwhile, proper digital quantum computing is still limited to about 10 qubits at a time, which scale-wise is kind of like a trainset inside a trainset inside of an actual train. So far no quantum error-correcting algorithm exists for real-life scales, but we're edging closer. The distance is still huge. "Before I die," Deutsch offered, "I would love to see just one universal logical qubit that can be indefinitely error corrected. It would instantly be classified by the government, of course. But I dream on, regardless."