JPMorgan Chase is investing in quantum computing technologies research to discover its potential uses for deep hedging within financial services. Deep hedging can be used to efficiently learn the expectations and distribution of returns, offer improved performance and train quantum policies.

The $3.6 trillion bank conducted a study last month to determine if deep hedging can reduce portfolio risk through data-driven models, according to a release from JPMorgan Chase and software provider QC Ware.

“The JPMorgan Chase team has performed experiments on a trapped-ion quantum processor,” Marco Pistoia, managing director, head of global technology applied research at JPMorgan Chase, told Bank Automation News.

The study found that deep hedging using quantum computing enables machine learning and AI models to be trained more efficiently. “Our research has demonstrated the potential of quantum computers to contribute to the solution of deep-hedging problems, paving the way for improved accuracy and trainability in the future,” Pistoia said.

Parallel processing and mitigating risk

“Quantum computing has the ability to run so many more situations at the same time,” BAN technology analyst Bradley Herbert said. “The benefit to the quantum is the more parallel processing and information.”

The technology can be used for risk management of a financial portfolio, Pistoia said, adding that “Quantum computing can help in this process by improving the accuracy and trainability of deep hedging models.”

Machine learning in the financial industry can be enhanced by quantum computing, specifically for analyzing time-series data, Iordanis Kerenidis, head of quantum algorithms at QC Ware, told BAN.

“One can leverage quantum computing for mitigating risks by looking not only at what will most probably happen in the future, but by basing decisions on the entire array of possible future outcomes,” Kerenidis said.