Johns Hopkins Team Breaks Through Quantum Noise

Newswise — Researchers from the Johns Hopkins Applied Physics Laboratory (APL) in Laurel, Maryland, and Johns Hopkins University in Baltimore have achieved a breakthrough in quantum noise characterization in quantum systems — a key step toward reliably managing errors in quantum computing. Their findings, published in Physical Review Letters, make important strides in addressing a long-standing obstacle to developing useful quantum computers.

Noise in quantum systems can come from traditional sources, like temperature swings, vibration, and electrical interference, as well as from atomic-level activity, like spin and magnetic fields, associated with quantum processing. Assessing the impact of noise on quantum algorithms is the first step to mitigating those effects, said Gregory Quiroz, a senior physicist at APL and an associate research professor in the Department of Physics and Astronomy at the Johns Hopkins University Krieger School of Arts and Sciences.

“Today’s models are commonly too simplistic to capture how quantum noise affects computation on real hardware,” Quiroz said. “Our work is trying to bridge that gap.”

A Matter of Time (and Space)

Many simplified models can only capture single instances of noise, isolated to one moment and one location in the quantum processor. But the most significant sources of noise spread across space and time, Quiroz explained.

“Capturing the effects of noise on the system over time and in multiple locations is really important to successfully implementing quantum error-correcting codes fault-tolerantly,” he said. “This is a problem we have to solve for large-scale quantum computers to work.”

Exploiting Symmetry

A quantum system becomes exponentially more complex as it scales up, making it even more difficult to understand how noise propagates in the system. To overcome this obstacle, Quiroz and co-author William Watkins, a physics graduate student pursuing his doctorate at Johns Hopkins within Quiroz’s research group, exploited a property of physics that helps simplify complex problems: symmetry.

“Symmetry provides structure, which allows us to simplify the problem by bringing in mathematical constructs that make it more tractable in the presence of noise,” Quiroz said.

Watkins realized that he could apply a mathematical technique called root space decomposition, a method that organizes how actions take place in a quantum system, to radically simplify how the system is represented and analyzed. The technique had been used to make progress in other areas of quantum mechanics, but to their knowledge, no one had applied it to quantum noise characterization before.

“It gave us insight into the problem in a mathematically compact and beautiful way, and gave us language to describe the problem,” Watkins said. “In one sense, you could say that our innovative framework is built on this mathematical foundation.”

Simply put, applying this technique allows a quantum system to be represented as a ladder, with each rung serving as a discrete state of the system. Quiroz and Watkins could then apply noise to the system to see whether specific types of noise caused the system to jump from one rung to another.

“That allows us to classify noise into two different categories, which tells us how to mitigate it,” explained Watkins. “If it causes the system to move from one rung to another, we can apply one technique; if it doesn’t, we apply another.”

This, in turn, will contribute in multiple ways to building error-resilient quantum systems, Quiroz said.

“Being able to characterize how noise impacts quantum systems helps us not only design better systems at the physical level but also develop algorithms and software that take quantum noise into account,” he said.

APL’s Quantum Portfolio

Quiroz noted that APL has expertise spanning the spectrum of quantum computing challenges — experimental physics, quantum algorithms, controlling quantum bits, and quantum error correction — and taking a noise-centric view of these research areas has been the main driver of the Laboratory’s work.

“Noise is a fundamentally hard problem standing in the way of large-scale quantum processors,” he said. “And APL is equipped with the expertise and ingenuity to solve it.”

“Our wide-ranging quantum noise portfolio includes studying fundamental sources of noise, such as cosmic rays, and developing novel noise characterization and mitigation protocols,” added Kevin Schultz, assistant program manager for Alternative Computing Paradigms in APL’s Research and Exploratory Development Mission Area. “We are very excited about this particular study due to the insight it provides on the impacts of noise on quantum algorithms and error correction, and we plan to pursue the potential research threads it suggests in the future.”