In September 2019, the claim leaked that a quantum computer developed by Google, codenamed Sycamore, had achieved “quantum supremacy” – the moment when quantum computers can execute calculations impossible with conventional machines. Sycamore was able to perform a calculation in three minutes and 20 seconds that it was claimed would take today’s most advanced classical computer, known as Summit, approximately 10,000 years.
Others have refuted the claim of “supremacy”, but Schrödinger’s cat does seem to be out of the bag, at least when we’re looking.
The particular calculation completed ‒ which essentially involved proving the randomness of a random number ‒ was extremely specific, but quantum computing is starting to emerge as a tool with broader applications. The Volkswagen (VW) Group has used quantum techniques to model traffic flows and researchers believe more sophisticated versions could remove the need for traffic lights. And Ford is working with NASA on ways quantum calculations can be used to solve problems around route efficiency, including how to reduce the climate impact of delivery vehicles.
"When quantum hardware becomes more reliable and available, risk management and pricing in derivatives and securities markets will be transformed."
A quantum of banking
This use of quantum computing to derive an optimal balance in a hugely complicated system with many possible variations or outcomes is of particular relevance to banking, and the increasing availability of access to quantum computers has not gone unnoticed by the more tech-savvy institutions.
So, in July 2019, Spain's CaixaBank joined forces with IBM to test the potential of quantum computing in the risk assessment of financial assets. The bank’s R&D department built two fictitious portfolios ‒ a mortgage portfolio and a treasury bill portfolio ‒ using real data to test the hypothesis that the use of a quantum algorithm to estimate credit risk would be more efficient than Monte Carlo simulations on classical computers.
More precisely, the quantum computer was used to estimate the economic capital requirement ‒ that is, the difference between the value at risk and the expected value of a given loss distribution. The economic capital requirement is an important risk metric because it summarizes the amount of capital required to remain solvent at a given confidence level.
Using IBM's open-source framework Qiskit ‒ an infrastructure that includes a simulator and a 16-qubit quantum computer ‒ the bank implemented a quantum algorithm and showed that while, ordinarily, the analysis requires thousands or millions of simulations, the quantum algorithm needed just dozens, cutting the time taken from several days to a few minutes. If portfolio complexity were increased, classical computers would be unable to perform the calculations in human time-scales, only a quantum system would be able to. This raises the possibility of cost-effective, dynamic portfolio optimization, currently a significant real-world problem for asset managers.
Unsurprisingly perhaps, quantum applications have also been found in one of the most complex areas of finance ‒ derivatives pricing. In July 2019, researchers from IBM presented a methodology “to price options and portfolios of options on a gate-based quantum computer… [providing] a quadratic speed-up compared to classical Monte Carlo models.” Translation: when quantum hardware becomes more reliable and available, risk management and pricing in derivatives and securities markets will be transformed.
"Sycamore was able to perform a calculation in three minutes and 20 seconds that it was claimed would take today’s most advanced classical computer approximately 10,000 years."
A quantum banking future?
These are just a few of the concrete, banking-related tests that have been run on real quantum computers in the last year or so. There are many more potential applications as the hardware, and the expertise in programming quantum algorithms, develops. Quantum computers could revolutionize machine and deep learning, allowing banks to search their vast datasets for patterns to detect and predict everything from potential misconduct through to customer preferences. Quantum algorithms have also been proposed for use in foreign exchange trading and arbitrage. The list goes on.
CaixaBank says it is now exploring the use of the technology applied to optimization algorithms, machine learning and secure encryption methods, but warns it is likely to be some time before the first commercial applications of quantum computing see the light.
That is surely true, but IBM’s ‘fleet’ of quantum computers accessible externally has reached 14, with a 53-qubit machine about to be added for clients of its IBM Q Network. The new system will be the largest universal quantum computer available for external use. These machines are far from having the thousands of stable qubits needed to create general purpose quantum computers, but hybrid conventional-quantum systems and “noisy intermediate-scale quantum” (NISQ) machines are already having an impact in chemistry, machine learning, materials science and cryptography. The quantum disruption of finance will be here sooner than people expect.