Prof. Dr. Andreas Mauthe
Scientific CV:
Dipl. Wirtsch.- Inf. (Mannheim, 1993), PhD (Lancaster University 1998), Research Associate (Lancaster University, 1994 - 1997), Chief Development Officer (Blue Order AG, Kaiserslautern, 1997 - 2002), Research Group Leader (KOM, TU-Darmstadt, 2002 - 2005), Senior Lecturer (Lancaster University, 2005 - 2013, Reader in Networked Systems (Lancaster University, 2013 - 2018), Professor in Networked Systems (2018), TU-Darmstadt Research Fellow, (TU Darmstadt, 2014 - 2020), Professor IT & Data Security (University of Koblenz-Landau, since 2019).
Research interests:
Security and resilience in networked systems (e.g. clouds, SCADA, smart grids, ICS, etc.). In his research, he has been working on the use of AI-based methods in anomaly detection and the analysis of network- and system-based datasets for about 10 years, building on this to develop mechanisms and methods for improving system resilience and security. This has recently evolved into research on the security and resilience of AI-based procedures themselves. This is essentially about how AI-based processes can be attacked (so-called adversarial attacks), e.g. through the manipulation of machine learning or the falsification of decision-relevant input data, and how AI processes and AI-based systems can be protected against this. Areas of application are, among others, in the area of network security or also email classification systems.
Security and resilience in networked systems (e.g. clouds, SCADA, smart grids, ICS, etc.). In his research, he has been working on the use of AI-based methods in anomaly detection and the analysis of network- and system-based datasets for about 10 years, building on this to develop mechanisms and methods for improving system resilience and security. This has recently evolved into research on the security and resilience of AI-based procedures themselves. This is essentially about how AI-based processes can be attacked (so-called adversarial attacks), e.g. through the manipulation of machine learning or the falsification of decision-relevant input data, and how AI processes and AI-based systems can be protected against this. Areas of application are, among others, in the area of network security or also email classification systems.
https://www.uni-koblenz-landau.de/de/koblenz/fb4/iwvi/agmauthe/team/prof-andreas-mauthe
