The first theoretical framework for a quantum computer was proposed in 1982 by Richard P. Feynman, and in less than 40 years, science and tech have rushed to build quantum machines. Today’s quantum computers sustain temperatures approaching absolute zero and are designed to solve problems that would require millions of years for even the world’s best supercomputers.
However, the rate of hardware development is seemingly outpacing the growth of algorithms that can leverage the phenomena of quantum mechanics.
Or to put it another way: “Everyone is trying to build these [quantum] machines, but we don’t know how to use them in many application domains,” says Helmut Katzgraber, a Principal Research Manager at Microsoft and an External Professor at SFI. “The number of quantum algorithms we have is limited, and most of them don’t really have any practical value,” he adds.
Quantum computers today excel at solving small toy problems for a select subset of disciplines, such as chemistry and physics, but the lack of practical algorithms limits their widespread application. And without useful algorithms, many fields will continue to rely on classical, silicon-based computers and potentially miss out on the revolutionary potential of quantum machines.
To address this shortage of algorithms, Katzgraber and his colleagues Maliheh Aramon (1QBit) and Jon Machta (the University of Massachusetts and SFI) are convening a working group at SFI July 30 through Aug. 2.
During the workshop, an interdisciplinary team of attendees is considering several themes posed as questions. They are considering topics that touch on which domains classical and quantum algorithms are likely to thrive, problems facing quantum computing, and recent developments in hardware, to name a few. The group will also discuss and develop algorithms for optimization, sampling, and machine learning.
“The main reason for the meeting is to think about the next generation of algorithms,” says Katzgraber. “We will not just focus on quantum hardware, but any type of hardware. We do not expect that a quantum device will be able to solve all problems; the key is to determine what problems will work really well, and what will not work at all.”
Katzgraber hopes the meeting will spur new algorithms, collaborations, and perhaps a new collection of white papers or a special issue of a journal.
Read more about the working group, New Algorithms for Optimization, Sampling, Learning, and Quantum Simulations