Oliwier Stefański from the Faculty of Applied Mathematics of the Silesian University of Technology took the first place in the competition “For the best diploma theses in the field of AI in Poland” organized by the Quantum AI Foundation for the work “Neuro-informatic biometric system”.

The main objective of this competition is to reward the authors of the best diploma theses and to create them appropriate conditions for further scientific and professional development in Poland. The competition and the foundation also place a strong emphasis on popularizing artificial intelligence, promoting outstanding academic achievements in this area and connecting talented graduates with leading experts and companies in the industry.

- The first place in the technical category was taken together with the late Artur Iwański, who unfortunately died shortly after his defence – says Oliwier Stefański. - Our work describes the implementation of the neuro informatic biometric system. It is a proprietary solution that allows verification of users based on brain activity. As it turns out, it is something completely individual for each person. Assuming the use of a simple and non-invasive sensor, our system can provide an alternative or support for classic biometric security. This solution can be particularly valuable for people with disabilities who already use assistive technologies, which do not always guarantee adequate security of their sensitive data.

The supervisor of the diploma thesis was Dr Eng. Adrian Kapczyński from the Faculty of Applied Mathematics.

- The diploma thesis, which I had the honour to be a promoter of, concerns extremely interesting issues in the field of biometrics, neuroinformatics, and artificial intelligence. This is an area of great potential, especially in the area of development of intelligent diagnostic systems, modern identification methods and systems supporting decision-making processes – says Dr Eng. Adrian Kapczyński. The work opens up broad perspectives for further research and development, including the use of deep learning methods in behavioural biometrics, securing systems against deepfake attacks, as well as the personalization of IT systems using neuroinformatics.

During the final gala, the student of the Silesian University of Technology had the opportunity to tell how the neural network prepared by them works, calculating the probability of convergence of the identity of the person logging in and how user data are classified, especially in the context of the alleged perception of components of human identity. The first place was taken ex aequo with the authors of two more works.

– An equally important element of the presentation was the issue that I consider fundamental in creating such systems – ethics. My little dream is that in the age of rapid innovation we do not lose human sensitivity. We should feel co-responsible for setting clear directions for the development of science, while at the same time focusing on minimizing the risk that this technology entails, - the laureate explains. - Together with us there were other great works on improving the quality of the SI models on the podium. The Gala brought together truly exceptional researchers solving real problems or pushing development opportunities even further, so I am very grateful that we were also able to find ourselves in such a group.

Congratulations on your success!

Photos by Maja Bodek