They are still largely products of the laboratory, but quantum computing applications may be reaching the point...
at which business leaders should begin to take notice.
Machine learning and AI are widely seen as fodder for future quantum approaches, but chemical simulation, cryptography and material science may be lining up first for the quantum treatment. In quantum computing application developments this week:
- Volkswagen at the CEBIT technology show in Hannover, Germany, disclosed some successes in its work with Google on quantum computers for battery research.
- Startup Strangeworks rolled out of stealth mode to describe plans to provide development tools for quantum computing aimed at the aerospace, pharmaceuticals, energy and finance industries.
- IBM announced ACQUA, for Algorithms and Circuits for Quantum Applications. When used with a previously available IBM quantum information science kit for software developers, ACQUA software is intended to allow domain experts in fields like chemistry, optimization and AI to run existing algorithms from classical computing jobs on IBM quantum computers on the IBM Cloud.
The quantum computing applications have evoked excitement even though they have yet to certifiably surpass the best of conventional computers; some call that inflection point "quantum supremacy."
The buzz has been triggered because quantum computing, based on murky-to-many atomic-scale quantum mechanics, could spur exponential increases in data processing. Someday, quantum approaches could blast past classical computers that support binary states of 0 and 1, by including additional states of 0 and 1, and 0 or 1.
Closer quantum view
Like other recent news on quantum computing applications, the work tends to be research-oriented. More than 20 years after it was demonstrated at the logical gate level, quantum computing still seems a ways off. But it may be getting closer.
Forrester analyst Brian Hopkins has a view on this. He has been tracking quantum computers for years, and admits it hasn't been so close that business leaders had to have it on their radar. That may be changing, though there is still some "fudge factor" in his time estimate, he said.
"We may be in that two- to three-year time frame -- or, definitely within five years -- where we are going to have some real uses for specialized non-error corrected quantum computers in certain industries," Hopkins said. "The leaders who are savvy enough to make the right investments could be in a position to reap first mover benefits."
Brian Hopkinsanalyst, Forrester Research
Elaborating on quantum computing applications, Hopkins said he divides quantum computers into specialized and universal systems.
The specialized variety can solve individual, practical problems. The universal type, like today's general purpose computers, is meant to handle all kinds of problems. Error correction in the quantum domain is somewhat different than error correction in digital circuits.
According to Hopkins, long-running jobs will stress quantum systems' abilities to maintain stable qubits. Error correction for quantum computing applications, he has written, is meant to yield smaller numbers of fault-tolerant, stable and logical qubits from many physical qubits.
Tech developers should learn quantum computing
Tech leaders will have to begin to learn some basics of quantum computing, in order to follow the future progress of different types of quantum computing, Hopkins maintained. That is especially true in science-intensive industries.
In those areas, he said, leaders need "to follow the progress of the different types of quantum computers to understand when they might achieve supremacy in the domains where they have a business problem that could benefit."
"For example, if you are in chemical manufacturing, and [researchers] hit supremacy in quantum chemistry -- if a Google or an IBM can create some quantum algorithms that run on a quantum computer that can solve some theoretical quantum chemistry problems -- you should know that," Hopkins said.
Hopkins recently reported on the status of quantum computing applications in a Forrester blog.
Vendor spins on quantum computing
- Google is working on a 72-qubit quantum chip that may prove the ability of quantum approaches to surpass classic computers in processing.
- IBM has built a 50-qubit processor, focused on advances in error correction and made quantum computing available to the public via the cloud.
- Microsoft has taken an approach based around unique Majorana fermion particles with hopes to reduce the circuitry need to correct errors in quantum computation.
Major tech companies getting into quantum
There is interest in machine learning, neural networks and AI among such quantum players as D-Wave Systems, Google, IBM and Microsoft. But Google, particularly, has AI on its mind when it comes to quantum endeavors, Hopkins said.
"The key to understanding why Google is doing this at all is to look at the name of its lab," he said. "They call it the Google Quantum AI Lab. The reason for that is they believe the primary application for quantum computing is accelerating artificial intelligence."
Quantum computing has a long history. With all the data needed to successfully train deep learning systems of the future, some exponential breakthroughs of the quantum kind could prove very helpful -- eventually.
As Hopkins and other quantum aficionados might say: Watch this space.