< !DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">
A century ago, the quantum revolution quietly began to change our lives. A deeper understanding of the behavior of matter and light at atomic and subatomic scales sparked a new field of science that would vastly change the world's technology landscape. Today, we rely upon the science of quantum mechanics for applications ranging from the Global Positioning System to magnetic resonance imaging to the transistor. The advent of quantum computers presages yet another new chapter in this story that will enable us to not only predict and improve chemical reactions and new materials and their properties, for example, but also to provide insights into the emergence of spacetime and our universe. Remarkably, these advances may begin to be realized in a few years.
From initial steps in the 1980s to today, science and defense agencies around the world have supported the basic research in quantum information science that enables advanced sensing, communication, and computational systems. Recent improvements in device performance and quantum bit (“qubit”) approaches show the possibility of moderate-scale quantum computers in the near future. This progress has focused the scientific community on, and engendered substantial new industrial investment for, developing machines that produce answers we cannot simulate even with the world's fastest supercomputer (currently the Summit supercomputer at the U.S. Department of Energy's Oak Ridge National Laboratory in Tennessee).
Achieving such quantum computational supremacy is a natural first goal. It turns out, however, that devising a classical computer to approximate quantum systems is sometimes good enough for the purposes of solving certain problems. Furthermore, most quantum devices have errors and produce correct results with a decreasing probability as problems become more complicated. Only with substantial math from quantum complexity theory can we actually separate “stupendously hard” problems to solve from just “really hard” ones. This separation of classical and quantum computation is typically described as approaching quantum supremacy. A device that demonstrates a separation may rightly deserve to be called the world's first quantum computer and will represent a leap forward for theoretical computer science and even for our understanding of the universe.
Once a real quantum computer is realized, what's next? In the coming decade, we can expect that some problem-solving will be optimized much more rapidly using quantum devices. We can also expect that efficient sampling from a probability distribution—the theoretical version of a machine learning algorithm—will become a place where quantum computers can shine. In the longer term, error correction and factoring may change the landscape further.
However, the lowest hanging fruit will be improving the ability to work with quantum mechanics. In the past, knowledge of quantum mechanics has been refined by comparing classical computational techniques to what has been observed by experiments—from solving differential equations to brute force simulation to new approximation methods in chemistry and materials science. If quantum computational supremacy is achieved, we may be able to test new techniques without requiring such comparison. This will reduce the cycle of research and transform how science is conducted.
Aiming toward these outcomes, the White House Office of Science and Technology Policy has established a new interagency group that is tasked with creating a national strategy to nurture a full quantum ecosystem through coordinated research between government, academia, and industry. This will include engagement across community boundaries and between disciplines to ensure a strong, quantum-smart future workforce. Matching this with similar efforts worldwide should allow us to catch glimmers of new scientific horizons and to develop the potential industries and new technologies that may emerge from investing in quantum information science.