Institut für Technik der Informationsverarbeitung (ITIV)
Julian Höfer, M. Sc.

Julian Höfer, M. Sc.

  • Engesserstr. 5

    76131 Karlsruhe

M. Sc. Julian Höfer

CV

  • Born in August 1996
  • Bachelor of Science Degree in Electrical Engineering and Information Technology at KIT
    • Graduated in February 2018
    • Bachelor Thesis: „Konzeption, Aufbau und Inbetriebnahme eines programmierbaren Steuerungsmoduls für eine Hybrid-Photovoltaikanlage“
  • Double Master’s Degree in Electrical Engineering and Information Technology at KIT and Grenoble INP Phelma
    • Graduated in February 2020
    • Master’s Thesis at ITIV: „Evaluation and Optimization of Highly Distributed Systolic Arrays for Neural Networks“
    • Internship at the European Synchrotron Radiation Facility (ESRF) in Grenoble (06.2018 – 09.2018): Work on a software extension for the embedded IcePAP motor control system
  • Employee at ITIV since April 2020

 

Research Interests

  • Hardware acceleration in the field of machine learning
  • Functional Safety of Artificial Intelligence in safety-critical environments

Artificial intelligence and neural networks have played an increasingly important role in recent years, especially in image processing. This progress enables highly complex applications like computer vision in autonomous driving. However, the more accurate these models become, the greater the computing effort. Since normal computers are no longer sufficient for this purpose, research is being conducted on special hardware accelerators that meet the demanding real-time requirements even when embedded in mobile systems.
In order to enable the use in safety-critical environments, research must be conducted not only on the „making safe“ of the algorithms, but also on the underlying new hardware architectures.

 

Teaching

Labor Schaltungsdesign (LSD)

 

Publikationen