Location History:
- Leuven, BE (2020)
- Heverlee, BE (2022 - 2023)
Company Filing History:
Years Active: 2020-2023
Title: Simon Johann Friedberger: Innovator in Machine Learning Security
Introduction
Simon Johann Friedberger is a notable inventor based in Heverlee, Belgium. He has made significant contributions to the field of machine learning, particularly in the area of model security and detection of copying. With a total of six patents to his name, Friedberger is recognized for his innovative approaches to safeguarding machine learning models.
Latest Patents
Friedberger's latest patents include a method for detecting if a machine learning model has been copied using intermediate outputs of the model. This method involves dividing the first machine learning model into multiple portions and comparing intermediate outputs from a hidden layer of a selected portion to corresponding outputs from a second model. If the outputs match, it indicates a high likelihood of copying. Another patent focuses on a method and data processing system for determining if a machine learning model has been copied, which includes adding an additional watermarking node to the output layer of the model. This innovation allows for the classification of input data while also providing a means to verify the originality of the model.
Career Highlights
Friedberger is currently employed at NXP B.V., where he continues to develop cutting-edge technologies in machine learning. His work is instrumental in enhancing the security and integrity of machine learning applications, making significant strides in the industry.
Collaborations
Friedberger collaborates with talented individuals such as Joppe Willem Bos and Nikita Veshchikov, contributing to a dynamic and innovative work environment. Their combined expertise fosters advancements in machine learning technologies.
Conclusion
Simon Johann Friedberger is a prominent figure in the realm of machine learning, with a focus on protecting intellectual property through innovative methods. His contributions are paving the way for more secure and reliable machine learning applications in various industries.