UC Berkeley to lead $25 million quantum computing center
As part of the federal government’s effort to speed the development of quantum computers, the National Science Foundation (NSF) has awarded the University of California, Berkeley, $25 million over five years to establish a multi-university institute focused on advancing quantum science and engineering and training a future workforce to build and use quantum computers.
The UC Berkeley-led center is one of three Quantum Leap Challenge Institutes (QLCI) announced today (Tuesday, July 21) by NSF and represents a $75 million investment. The initiatives are a central part of the National Quantum Initiative Act of 2018, the White House’s Industries of the Future program and NSF’s ongoing Quantum Leap effort.
The QLCI for Present and Future Quantum Computation connects UC Berkeley, UCLA, UC Santa Barbara and five other universities around the nation, harnessing a wealth of experimental and theoretical quantum scientists to improve and determine how best to use today’s rudimentary quantum computers, most of them built by private industry or government labs. The goal, ultimately, is to make quantum computers as common as the mobile phones, which are digital computers, in our pockets.
“There is a sense that we are on the precipice of a really big move toward quantum computing,” said Dan Stamper-Kurn, UC Berkeley professor of physics and director of the institute. “We think that the development of the quantum computer will be a real scientific revolution, the defining scientific challenge of the moment, especially if you think about the fact that the computer plays a central role in just about everything society does. If you have a chance to revolutionize what a computer is, then you revolutionize just about everything else.”
Situated near the heart of today’s computer industry, Silicon Valley, and at major California universities and national labs, “this center establishes California as the world center for research in quantum computing,” he said.
Quantum computers are fundamentally different from the digital computers in our cellphones, laptops, cars and appliances. You can think of digital computers as a collection of millions of independent bits — either ones or zeros — that flip back and forth every billionth of a second based on a series of instructions called an algorithm. The harder the problem, the longer the list of instructions.
In a quantum computer, each bit is linked to every other bit — quantumly entangled — so that the description of the state of even 100 quantum bits would be far larger than could be stored on the biggest classical digital computer.
“Translating this remarkable ability of quantum computers into actually solving a computational problem is very challenging and requires a completely new way of thinking about algorithms,” said Umesh Vazirani, UC Berkeley professor of computer science and co-director of the institute. “Designing effective quantum algorithms is a key challenge in realizing the enormous potential of quantum computers.”
Theoretical work has shown that quantum computers are the best way to do some important tasks: factoring large numbers, encrypting or decrypting data, searching databases or finding optimal solutions for problems. Using quantum mechanical principles to process information offers an enormous speedup over the time it takes to solve many computational problems on current digital computers.
“Scientific problems that would take the age of the universe to solve on a standard computer potentially could take only a few minutes on a quantum computer,” said Eric Hudson, UCLA professor of physics and co-director of the new institute. “We may get the ability to design new pharmaceuticals to fight diseases on a quantum computer instead of in a laboratory. Learning the structure of molecules and designing effective drugs, each of which has thousands of atoms, are inherently quantum challenges. A quantum computer potentially could calculate the structure of molecules and how molecules react and behave.”
“I think quantum computing is inevitable,” Stamper-Kurn added. “I don’t know the time scale — Is it 100 years or 10 years? — but we are talking about exponential increases in capability.”
The scaling problem
At the moment, quantum computers typically yoke together a paltry 50 or fewer quantum bits, or qubits. But that is quite an achievement, Stamper-Kurn said, considering that it came about rapidly over the past decade and has already spawned a nascent quantum computer industry. In light of these advances, scientists and the federal government anticipate even faster progress if the government invests in basic research and education that complement the technical progress being made by companies such as Google, Microsoft Corp., Intel and IBM.
The new institute, which also includes the University of Southern California, California Institute of Technology, University of Texas at Austin, Massachusetts Institute of Technology and University of Washington, Seattle, will tackle some of the major challenges in the field.
“We know that quantum computers are on their way — there are researchers across the country working to build and test them,” said NSF program director Henry Warchall. “But anyone with a computer will tell you that hardware isn’t useful without software to run on it — and that’s where this center will lead us toward solutions. The NSF Quantum Leap Challenge Institute for Present and Future Quantum Computing will help us have key programming elements in place when quantum computing hardware is in place.”
One of the institute’s first challenges is to identify the applications for which current quantum computers are most suited, in order to make full use of today’s first generation computers.
“People talk about noisy intermediate scale quantum computers, or NISQ devices, which is what we have at present. They are pretty limited in what they can do, most importantly because they don’t know how to correct the errors that come up during the computation,” Stamper-Kurn said. “They are going to be useful for short-scale or small-scale computation, but it is critical that we find ways to use them productively, because that will stimulate the whole field.”
The institute will also address the long-term challenge of developing algorithms for the next generation of quantum computers that will enable critical scientific, economic and societal advances, as well as navigate the boundary between quantum and classical computational capabilities.
“Realizing the full power of quantum computation requires development of efficient schemes for correction of errors during operation of quantum machines, as well as protocols for testing and benchmarking,” Vazirani said.
Understanding the computational capabilities of quantum computers is one of the most important challenges for the field and will be an important driver of progress moving forward. This will require a major increase in the number of computer scientists engaged in these questions.
“The Simons Institute for the Theory of Computing at UC Berkeley is uniquely poised to create this engagement,” said Vazirani, who is leading the quantum computing effort at the institute. “The Simons Institute is a mecca for the foundations of computing and will host a number of researchers in quantum computing and facilitate the kind of intense, in-person, cross-disciplinary collaboration that can result in rapid progress.”
Also key is a partnership with UCLA’s Institute for Pure and Applied Mathematics, which will help apply mathematical and data science tools to the field.
The magnitude of the challenge also requires input of domain expertise from scientific and mathematical/computational disciplines, to allow quantum algorithm design to be tailored to specific problems.
“Quantum algorithm design is entering an era of co-design, where the specific scientific and computational constraints and the need to preserve the fragile quantum coherence underlying quantum algorithms are leveraged to generate an efficient solution to a particular scientific problem,” said Birgitta Whaley, UC Berkeley professor of chemistry and co-director of the institute. “We know how to do this for small systems, but scaling up to large quantum machines brings new challenges for implementation. That is something we will tackle in the new institute.”
Experimentalists and theorists from the fields of chemistry, physics, materials science, engineering, mathematics and computer science will tackle some of these outstanding problems — in particular, how to scale up computers from tens to millions of qubits without losing the quantum properties of the ensemble of qubits.
“The big question is: How do you make a quantum system bigger and bigger without making it perform worse and worse?” Stamper-Kurn said. “What people see is that, as the thing gets bigger, more noise creeps in, calibration is more difficult, connectivity is difficult — it is hard to get one part of the computer to talk to the other.”
The group plans to focus on three experimental platforms that use different quantum systems as qubits: trapped ions, trapped atoms and superconducting circuits.
“Some of these systems work great at a small number of qubits, so we can work really hard on increasing their fidelity, making them more accurate. Some naturally operate at a larger number of qubits, and we can test out ideas about how to operate a quantum computer with a limited range of controls, but a lot of qubits to work with,” Stamper-Kurn said. “They are all early in the technological development curve, so by introducing some new technologies with the help of engineers, we can improve our ability to operate a lot of systems at once.”
The grant will foster interactions among researchers and doctoral students from many fields with the help of fellowships, conferences and workshops. But a major component will be training a future workforce akin to the way computer science training at universities like UC Berkeley and Stanford fueled Silicon Valley’s rise to become a tech giant. UCLA will pilot a master’s degree program in quantum science and technology to train a quantum-smart workforce, while massive online courses, or MOOCs, will help spread knowledge and understanding of quantum computers even for high school students.
The team hopes to partner with Department of Energy laboratories, such as Lawrence Berkeley National Laboratory, which in 2018 launched an Advanced Quantum Testbed to further quantum computation based on superconducting circuits.
The project came to fruition, in part, thanks to a UC-wide consortium, the California Institute for Quantum Entanglement, funded by UC’s Multicampus Research Programs and Initiatives (MRPI).
“The award recognizes the team’s vision of how advances in computational quantum science can reveal new fundamental understanding of phenomena at the tiniest length-scale that can benefit innovations in artificial intelligence, medicine, engineering and more,” said Theresa Maldonado, UC’s vice president for research and innovation. “We are proud to lead the nation in engaging excellent students from diverse backgrounds into this field of study.”
Co-directors of the institute are UCLA’s Eric Hudson; Whaley, who is co-director of the Berkeley Quantum Information & Computation Center (QBIC); Vazirani, the Roger A. Strauch Professor of Electrical Engineering and Computer Sciences and co-director of the Berkeley Quantum Information & Computation Center (QBIC); and Hartmut Häffner, UC Berkeley associate professor and the Mike Gyorgy Chair in Physics.
The two other $25 million Quantum Leap Challenge Institutes announced today are centered at the University of Colorado, Boulder, and the University of Illinois, Urbana-Champaign, and will focus on quantum sensing and quantum networks, respectively.