CQE PI Feature – Kevin M.Obenland

Featured in QSEC July newsletter 2024

What would we do with a large-scale quantum computer if/when we are able to build it? This basic question motivates much of the research in my group. Significant progress on quantum hardware has been made in the past 10 years. However, if we don’t have applications that demonstrate real utility, then the impact of these platforms will not be as significant as we hope. Our group is investigating applications in physical science where a quantum computer has measured advantage over a classical computer.

In evaluating applications for potential utility on quantum computers, one can approach the problem from two directions.  Looking top-down, one must consider how well an application maps to a quantum representation and how well it utilizes the quantum algorithms that we currently have at our disposal. Quantum chemistry is one area where quantum computers may provide an advantage. Classical implementations of many of these problems requires exponential space and time, which is reduced to polynomial on a quantum computer. Additionally, many of the things that we want to do with these systems, for example finding ground states and measuring observables of quantum dynamics, have well established quantum algorithms.  Looking bottom-up, one can ask what resources are required to solve a problem of a specific size. How many qubits are required and how long will the computation take? As part of DARPA’s Quantum Benchmarking program, we have been helping to answer these questions. We have developed an open-source software package (called pyLIQTR) that contains implementations of high-utility quantum applications which can be decomposed to logical gates. Our tool provides a high-level count of resources or can be fed to physical resource estimation tools like the Azure quantum resource estimator, Bench-Q, or Resource Estimator from Rigetti.

My introduction to quantum computing came as a happy accident. While l was looking for a thesis topic at the University of Southern California, my Ph.D. advisor, Alvin Despain, participated in a JASON study focused on the new field of quantum computing. The JASONs were a group of senior academics and scientists that met in La Jolla California each summer to study various scientific questions for the US government. In the summer of 1995, the topic was Quantum Computing. As a result of this study my advisor and I were introduced to many of the scientists that would become leaders in quantum computing including Jeff Kimble, John Preskill, Ike Chuang, and Seth Lloyd, and we became involved in a DARPA program investigating alternative platforms for computing.

And so, it was decided that I would work on this program and pursue quantum computing as my thesis topic. My initial work involved building a classical simulator of a quantum computer which was used to understand the impact of error on a quantum computation. My simulator was the first of its kind, and, perhaps because it was so ahead of its time, this is the reason that my work doesn’t get referenced in current research on quantum simulators.

Our current quantum computing resource estimates for utility-scaled problems in physical science are quite high. I believe, as has been the case for classical computing, the community will continue to develop new algorithms, will optimize the methods that we have, and will discover new application areas where quantum computing can have an impact.