PhD position: Security of XAI/ XAI for IT-Security

The position is fully funded with the German salary level TV-L 13 (100%) and should be filled at the soonest possible date. The position offers the possibility of funding the entire duration of the PhD.

Application up to: June 15th, 2023

Start: In the course of 2023

Research

The "Intelligent System Security" research group works at the intersection of machine learning and computer security. We develop learning-based methods for attack detection on different levels or the discovery of vulnerabilities in software. Moreover, we research the robustness and security of machine learning methods themselves.

Your Profile

You have a Diploma or Master's degree in computer science or any related field. You also require very good knowledge in machine learning and ideally computer security. Above all, however, you need to show enthusiasm for conducting research on cutting-edge topics in machine learning and computer security.

Field of Work

Explainable machine learning (XAI) can be very well used for attack detection in both focus areas of the research group (Security of ML and ML for IT-Security). For instance,

Detection and prevention of attacks against computer systems (MLSEC)
  • Analysis of attacks and malware
  • Assisted discovery of vulnerabilities and fuzz testing (Fuzzing)
Detection of attacks against machine learning (SECML)
  • ML backdoors
  • Adversarial examples
  • Model stealing

Moreover, the robustness against attacks on the XAI methods themselves is crucially important. Next to demonstrating new attacks, in particular, developing robust XAI methods is essential. For instance,

  • Input manipulations or model manipulations to disguise attacks
  • Fairwashing XAI results
  • Certifiable robust XAI methods

Community

All Ph.D. students will be members of the KIT Graduate School Cyber Security.

Application

Please send your application including a cover letter, your CV, certificates/references, and a research proposal to . Make sure to point out why you are a good fit for us and research in computer security.

We strive for a diverse research team and ask people from all nations to join us. Moreover, qualified women are strongly encouraged to apply. Disabled persons with equivalent aptitude will be favored.