Deutsche Börse relies on quantum computers

A global DAX group like Deutsche Börse is exposed to many risks. In future, it should be possible to calculate this in half an hour instead of decades.

IBM-Quantencomputer der Serie Q System One

Dhe Deutsche Börse sees good opportunities for the practical application of quantum computers in a few years. That emerges from the results of a pilot project, which will be published this Wednesday. Accordingly, the calculation time of complex models can be drastically reduced. “We wanted to know whether quantum computing can help us to calculate the risks of our business activities faster and better and in what time horizon this could be used,” explains Michael Girg, responsible for the cloud business at Deutsche Börse.

Companies have also been using risk models to date. “But we have hundreds of variables in macroeconomic models, such as changes in interest rates, trade disputes and much more, and we want to know how much our risk changes if we vary the assumptions for future developments,” says Girg. “Calculating these parameters in such a level of detail with classic computers would take decades. That is impractical. We need the results after 24 hours at the latest. “

The stock exchange’s project partner is Jos Quantum, a Frankfurt fintech founded three years ago. “What was special about this project was not only that we developed a new quantum algorithm, namely the sensitivity analysis, but that it can be used for a very practical problem in the financial world,” says Niklas Hegemann, one of the managing directors of Jos Quantum. Result: The computing time of the simulation drops with quantum computers from a decade to 30 minutes. Not yet today, but perhaps in 2025, assuming that the billions in public funding that are currently being invested in the further development and construction of quantum computers will lead to corresponding progress.

Calculate more states than atoms in the universe

“This is an important insight for us to go further in this direction,” says Girg. A next use case is to be tested this year. Jos Quantum has focused on applications in finance, insurance and energy. A model is currently being developed in the energy sector. which is funded by the federal government as part of a research program.

However, models for credit risks or systemic risks or the area of ​​optimization are also conceivable. Here, quantum computers can try out solutions in parallel in complex models, which would take years with classic computers. “As physicists and computer scientists, we are fascinated by the fact that only 300 quantum mechanical bits can be used to calculate more states at the same time than there are atoms in the universe. We now want to use this potential, ”says Hegemann.

The acceleration of computing processes raises high hopes for companies like the stock exchange. “We are satisfied with the robustness of current models, but to be able to assess risks even better, vary models even more finely and to be able to simulate and calculate the influences of parameters on each other much better, would enable a significantly better overall risk assessment through quantum computing,” says Girg. The current study can be found under the link: https://arxiv.org/abs/2103.05475