2021 Year in Review
Building the Quantum Advantage Evaluation Framework
Quantum computing offers a possible solution for hard computing problems in the Department of Defense (DoD), such as materials science and combinatorial optimization. But this nascent technology requires significant, targeted investment. To help guide the DoD’s research and development, the SEI developed the Quantum Advantage Evaluation Framework (QAEF), which will be used to predict the computing applications that will show quantum advantage in the next one to three years.
“Quantum advantage occurs when quantum computing can solve some practical DoD problem faster, or create a higher quality of solution, or both, than alternatives like classical state-of-the-art computing,” said Jason Larkin, an SEI senior researcher leading the QAEF project.
QAEF will compare application benchmarks on both quantum and classical state-of-the-art computing to determine quantum advantage. QAEF will be able to benchmark any DoD-relevant applications and algorithms by executing them on real and simulated quantum devices. “That’s challenging because to get optimal performance, as in classical computing, you’ve got to optimize for quantum computing across the full stack, which is currently under development,” Larkin said. Though stack development may take years, Larkin and his team have already begun comparing application benchmarks.
Quantum advantage occurs when quantum computing can solve some practical DoD problem faster, or create a higher quality of solution, or both, than alternatives like classical state-of-the-art computing.Jason Larkin
Senior Researcher, SEI AI Division
The SEI has expertise in the algorithms and software engineering of both quantum and classical computers. The SEI collaborated with QuantumHub at Carnegie Mellon University (CMU); CMU researchers; government organizations, such as the research laboratories of the Air Force, Army, and Navy; and the Pittsburgh Supercomputing Center. The SEI also recently became part of the IBM Quantum Network and in the past year has transitioned its fundamental research into customer contexts.