Quantum computing is receiving a fair amount of hype at the moment. Hundreds of millions of dollars are being poured into research, from both the private sector as well as business, with the belief that this really is the next big thing.
But how close are we, in reality?
At Global Derivatives 2017 in Barcelona the very people at the heart of recent developments in the quantum field told us just how close we were to a world in which quantum computing would become the norm.
The reason everybody wants to know, of course, is that quantum computing could be a real game-changer.
What is it, though?
What is quantum computing?
Professor Helmut Katzgraber from the Department of Physics & Astronomy at Texas A&M University, took a good stab at stealing Justin Trudeau’s crown for best concise explanation of quantum computing when he said: “Input goes in, then magic happens.”
Vasil Denchev’s explanation illuminated why quantum computing was so important:
“There is a lot of evidence that quantum computers will be able to solve certain computational tasks much faster than classical computers. This could have very practical outcomes for different areas of science and business,” he theorised.
As Software Engineer at the Quantum Artificial Intelligence Lab at Google, Vasil’s day job involves making this theory a reality.
What are the issues surrounding quantum computing?
Quantum computing is not without its problems, though.
“We do have optimisation problems in building quantum computers,” stated Davide Venturelli, Science Operations Manager at Quantum Artificial Intelligence Laboratory, NASA Ames Research Center. “Perhaps one of the early applications of a quantum computer should be to build the quantum computer,” he reasoned.
Different sectors, including finance, could contribute to finding a solution to those optimisation problems, he added.
Helmut pointed out that claims of the quantum computer’s super speed had to be taken with a grain of salt.
“When you talk about speed up there are two important ingredients,” he explained. “One, how much faster is it? And two, how does this speed change with a number of variables?”
So far, experiments involving quantum computing have focussed on the D-Wave machine – the only available quantum optimiser that implements a restricted version of quantum computing.
Vasil explained that his experiments had showed that, for a certain crafted computation problem that fits very well on the D-wave machine, they could achieve a speed up to 100 million times faster than classical computers.
But this came with the caveat that this was a crafted problem, and the purpose of the experiment was just to show that this was possible, rather than any practical implication.
“We haven’t been able to do this for a more practical problem yet, but it’s important to know that speed like that is possible.”
What is the near-term impact of quantum computing?
What might be the first impacts of quantum computing on the finance world?
In the fintech space, for instance, would this enhanced computational power make cryptographic protocols, such as those used in blockchain, less safe?
No, said Helmut. “It’s going to take a while before we have a machine that can implement algorithms that can factor large numbers. This will come, but not in the near future. “Having said that, if it happens, there is quantum encryption.”
Helmut felt that only when digital devices overtook analogue would we see a real impact, not just in science but in applications in industry.
“An analogue machine is a device that has intrinsic errors, biases and noise,” he explained. “So there is only so much that you can really do. There are many optimisation problems that require high precision, and this precision is not available in analogue,” he explained.
Google was investing in both analogue and digital: “Because we can,” said Vasil. “But also because we don’t know for sure which one will have the largest pay off in the short term.”
“We still have a lot of hope for the analogue approach for quantum annealing. Our current thinking is that a hybrid quantum/classical optimisation approach might be the most successful because that allows us to get the best of both worlds.”
20 years from now, what percentage of the world will have quantum computing?
The use of quantum computing may remain a specialised area for quite some time, felt the panel members.
Helmut explained how there were two obstacles to the proliferation of quantum computers.
“First, will we have large enough digital devices that can do things that are useful? Second, there is the fact that there are not too many quantum algorithms out there.
“In my opinion, there will be small pockets this special hardware will be used, but where it is used, there will be a very large impact.”
Vasil agreed: “I don’t envision having general purpose quantum computing that can be used for mundane tasks.
“But I might be wrong,” he said.
And even a two to three percent improvement could be game-changing, interjected Davide.
“There is so much that we need to learn about quantum information science,” admitted Davide. “But the number of scientists is growing, and the investment in it is growing.”
It may be here sooner than we think.