Leo Buron

Power-Efficient Spike Sorting for Brain Computer Interfaces / Online Learning for Embedded Hardware Accelerators

Bio:

Leo Buron, M.Sc, has been working as a research assistant and doctoral student at the Department of Embedded Systems since May 2022. After completing his dual studies in electrical engineering at the University of Applied Sciences in Düsseldorf in cooperation with Siemens, he moved to the University of Duisburg-Essen for his master’s degree. There, he studied Electrical Engineering and Information Technology with a focus on embedded systems and worked part-time at Siemens Energy. His master thesis was about the efficient design of LUTs on FPGAs for precomputed neural networks.

Description of the Talk:

In most biomedical applications a fine tuning for each patient is necessary. In addition, for some applications adaptation through runtime are needed. Therefore, this research presents an approach for runtime adaptation for next generation BCIs. Thus, the SOTA of spike sorting is presented and adaptation mechanisms are explained.