news

DARPA Awards $9.2M Grant to Inter-agency Team Researching Quantum Computing

Primary tabs

Quantum computing is a young, exciting field in computer science. Experts hope that its powerful processing capabilities will eventually help the world solve increasingly complex problems.

To explore these new possibilities, the U.S. Defense Advanced Research Projects Agency (DARPA) has awarded several high-value, multi-stage grants. One such grant has gone to inter-agency group OPTIQ (Optimization with Trapped Ion Qubits). Led by Georgia Tech Research Institute (GTRI), OPTIQ is a collaboration with Georgia Tech, Oak Ridge National Labs (ORNL), and the National Institute of Standards and Technology (NIST). Creston Herold, senior research scientist in GTRI’s quantum systems division and head of the measurement branch, is the lead PI. Swati Gupta, an assistant professor in the H. Milton Stewart School of Industrial and Systems Engineering, is the Georgia Tech PI, along with John Bollinger (group Leader of NIST’s ion storage group) and Travis Humble (distinguished scientist and director of the ORNL Quantum Computing Institute), who round out the team. DARPA has given OPTIQ up to $9.2 million to do its work over a four-year period.

“Our OPTIQ team is going to explore special-purpose quantum hardware tailored to solving combinatorial problems like Max-Cut, as well as to show what’s called ‘quantum advantage,’ – i.e., quantifiable advantage over specific instances of such problems where quantum hardware can outperform classical algorithms,” explained Gupta. Max-Cut is one of the most famous NP-hard problems that arises in many operations research applications. Researchers have been attempting to solve it for over five decades using classical – or digital – computers.

On the faculty in one of the foremost optimization departments in the country, Gupta is particularly interested in examining how challenging problems in classical optimization can be solved by quantum computing.

“This program is really a race between classical and quantum models of computation. This means that we need to identify blind spots, i.e., identify the really hard problem instances, for classical methods that can be improved by current quantum hardware,” she said. “Imagine there are two superheroes: one (classical) that is lightning fast but runs on the ground, while the other (quantum) can jump to arbitrary far-off spaces as long as these spaces have a landing pad. There might exist some places that the second superhero can get to faster than the first one. We need to find these instances. This will give us an opportunity to find barriers for classical optimization and develop a deeper understanding of what makes problems solvable classically.”

She continued, “What particularly excites me about this project is the opportunity to think beyond the classical model of computation that has paved the way for algorithmic design for decades, and to look for new computational primitives that might change the face of computation, as well as train a new generation of researchers who are adept at both classical and quantum techniques.”

The OPTIQ team will build quantum hardware based on trapped-ions to run combinatorial optimization problems, like Max-Cut, and benchmark its performance against the best “classical” optimizers. GTRI and NIST will collaborate on quantum hardware design, construction, and operation. GTRI and ORNL will develop methods to learn the best operating parameters for the quantum hardware and mitigate noise. Georgia Tech will identify “hard” instances of Max-Cut where quantum hardware may have a quantifiable advantage over known classical heuristics. The OPTIQ project leverages ORNL’s expertise in large-scale modeling of noisy quantum systems.

Quantum computing has promising applications in pharmaceutical discovery and faster scientific simulations, as well as revolutionizing artificial intelligence and the search for new materials. The OPTIQ team is looking forward to the potential insights that may be generated by bringing together quantum physicists and optimization experts.

Status

  • Workflow Status:Published
  • Created By:Shelley Wunder-Smith
  • Created:06/15/2020
  • Modified By:Shelley Wunder-Smith
  • Modified:06/15/2020

Categories

  • No categories were selected.