Architecture and Hardware

Addressing Quantum Error Correction

A centimeter-sized silicon chip, which has two parallel superconducting resonators and quantum-circuit refrigerators connected to them.
A new technique could reduce or eliminate the need for quantum error correction.

Arguably quantum computing's biggest challenge is error correction, since qubits—the quantum equivalents of digital ones and zeros—can prematurely collapse into an ordinary digital bit during a quantum calculation.

In quantum calculations, qubits superimpose values of one and/or zero to speed up a calculation, collapsing into the answer as the last step. Most of the efforts of quantum researchers so far have been on how to extend coherence as long as possible, a noble endeavor but not the whole story, according to Mikko Möttönen, senior scientist and group leader of the Center for Quantum Engineering at Finland's Aalto University, who worked on this with Matti Silveri of Finland's University of Oulu. These researchers and collaborators are focused on eliminating one of the primal sources of quantum errors: by controlling the length of coherence and precise time of decoherence, quantum calculations become more reliable, decreasing the need for error-correction circuits that fix the problem of premature decoherence.

Said Stefan Filipp, technical leader of Superconducting Qubit Quantum Computation at IBM Research–Zurich, "The beauty of quantum information processing is the exquisite control that has been gained over complex quantum systems in the last decades. Dr. Möttönnen and his team at the Aalto Center for Quantum Engineering have successfully demonstrated how to accurately control not only single quantum circuits but also their environment. With this extra control knob, the amount of dissipation and coherence can be manipulated to study the role of the environment in quantum processors and helping to make them quicker."

Möttönen and associates adjust the coherence time to the length of a particular quantum operation, always guaranteeing the operation does not exceed the capabilities of the environment around the quantum gates in use, thus avoiding errors from premature decoherence. After the operation is completed, the new method quickly resets the qubits so the next quantum calculation can be started. Möttönen believes this method will dramatically speed up quantum computers, as well as eliminating the need for time-consuming error-correction checks.

"We want the maximum coherence time to be as long as possible, but our way of making the coherence 'lifetime' to be as short as possible for each calculation gives some slack to this requirement," said Möttönen. "In a fault-tolerant quantum computer, where one needs to measure, operate, and initialize qubits at every cycle, the initialization speed compared to the maximum coherence time has to be fast. Our method makes this ratio big, and hence can help in building quantum computers."

The method developed by Möttönen et al. works by closely controlling the energy losses associated with quantum gate operations, which means it applies directly to the "universal" quantum computers built by Google, IBM, Microsoft, Rigetti, and Xanadu—but not the quantum annealers made by Canada's D-Wave Systems and emulated by Fujitsu.

"Energy losses and decoherence are intimately related. More precisely, the rate for loss of coherence is, at maximum, half the rate of energy loss," explained Möttönen. "Since the energy losses restrict coherence, we of course want to reduce them…in brief, one should induce losses once one wants to reset the qubit, and have no losses during the quantum logic operation."

The researchers do this by adding a normal metal-insulator-superconductor tunnel junction, which can be switched on to induce decoherence and reset a circuit's quantum gates.  This induced decoherence method, capable of quickly resetting quantum gates, was demonstrated to be tunable over a wide range, allowing specific quantum operations to be precisely controlled to minimize premature quantum decoherence and the need for error-correction circuitry.

"We can control when the coherence time is long and when it is short," said Möttönen. "It is amazingly efficient. We can have it as small as we want to begin with, and then increase it more than a thousandfold when we want."

In experimental research at the Aalto Quantum Computing and Devices research lab, Silveri focused on the quantum simulations and theoretical design of the fabricated quantum circuits. According to Silveri, the key to their success was the observation that energy dissipation can be turned on and off, and that the control of such energy shifts is critical for the implementation of gate-based quantum computers.

Silveri traces their discovery back to Nobel Laureate Willis Lamb, who made the first observations of stunning changes from small energy shifts in hydrogen atoms over 70 years ago. According to Silveri, he and Möttönen engineered the same physics into quantum systems for the first time, a development that will be critical to the successful implementation of universal quantum computers. Called the Lamb shift, it was induced for the first time in their engineered quantum system.

Said Rogério de Sousa, an assistant professor and leader of the Condensed Matter Theory group at the University of Victoria, Canada, "It is indeed a breakthrough in quantum engineering. Silveri et al designed a quantum circuit coupled to a normal metal island. You can think of it as a single photon mode coupled to a dissipative heat of electrons. By controlling the bias voltage, they show that the interaction between the photon 'cavity mode' and the electron environment changes by more than a factor of 1,000. This dramatic tunability allowed them to resolve one of the most conspicuous effects in quantum physics: the energy shift of a quantum system due to its interaction to a bath (or 'environment'), the so-called Lamb shift."

With optimized sample parameters, the researchers hope their technique will allow the systematic study of the Lamb shift in recently synthesized ultra-strong coupled artificial atoms—part matter, part light—which exist in a unique non-perturbative quantum state.

De Sousa added, "Their 'tunable quantum bath' will open new avenues on the research of open quantum systems, such as measuring the Lamb shift in quantum systems in the strong coupling regime, where the usual theoretical methods break down. If they are able to demonstrate this effect in resonators with quality factors of 106 [one million] or more (such as those used in the IBM and Google quantum computers), this will become a useful tool for quantum computing."

R. Colin Johnson is a Kyoto Prize Fellow who ​​has worked as a technology journalist ​for two decades.

Join the Discussion (0)

Become a Member or Sign In to Post a Comment

The Latest from CACM

Shape the Future of Computing

ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.

Get Involved

Communications of the ACM (CACM) is now a fully Open Access publication.

By opening CACM to the world, we hope to increase engagement among the broader computer science community and encourage non-members to discover the rich resources ACM has to offer.

Learn More