In collaboration with the consulting firm KPMG, a student from DTU Physics has tested whether a quantum computer is better at optimizing a stock portfolio than a conventional computer.
During his Master of Science in Engineering programme, Mathias Hoeg Boisen has had several courses in both quantum informatics and quantum mechanics. The area interests him, but he has also experienced a growing sense that the knowledge he acquired on the university courses was far from being used in the real world.
Therefore, Mathias approached one of his lecturers to inquire whether it was possible to collaborate with a company on a specific quantum technology project. It did not take long before contact was established with the consulting firm KPMG, which already has an eye on quantum computers as part of its future technological toolbox.
The next step was to find a problem that had a sufficient scope for it to be interesting to solve using a quantum computer. Here, the choice fell on stock portfolio optimization, which is a classic problem in the financial sector. If you have a given budget and want to protect yourself against heavy losses, which shares should you then choose to buy to achieve the best financial result?
Better solution, but
In specific terms, Mathias used a type of quantum computer known as quantum annealer, developed by the company D-Wave.
The stock portfolio was set at 65 shares, which is characterized as a big assignment because it offers a vast number of different buying and selling options. The quantum computer was then used for the calculations, and the result was subsequently compared with classic methods. It quickly turned out that the quantum computer could beat the simplest classic method—brute force—and found the most optimal solution for up to 25 shares.
“However, it was clear that the quantum computer still has some challenges in terms of having sufficient capacity and performance to outperform other classic methods for really large-scale calculations with more than 25 shares,” says Mathias.
“Since our experiments, however, D-Wave has created a new and more powerful processor that performs better. With this development, the quantum computer will probably overtake classic methods completely over the next couple of years. This will only increase the interest banks have already shown in our work.”
Other students can do the same
Mathias is currently completing the final part of his study programme, an MSc thesis on another quantum computing model. He subsequently hopes to use his knowledge, for example in a company.
“I’m happy to have worked so closely with a company during my studies. I performed half of my project work at KPMG and the other half at DTU, which has given me a really good insight into the company’s practical application of quantum computer technology,” he says.
The great openness to the performance of a project that matched his wishes makes Mathias encourage other students to explore the same opportunities for landing their dream assignment. Also even if you have a slightly offbeat or narrow interest like himself, because he feels that—at DTU—there is a willingness to find a solution that includes both new types of business collaborations and the inclusion of real-world issues.