Christoph R. Landolt
ELLIS PhD Student at CISPA Helmholtz Center for Information Security
Saarbrücken, Germany
I’m a doctoral researcher at the CISPA Helmholtz Center for Information Security in Saarbrücken, Germany, supervised by Mario Fritz (primary advisor, CISPA) and Florian Tramèr (secondary advisor, ETH Zürich — Secure and Private AI (SPY) Lab). As an ELLIS PhD student, my research focuses on agentic and self-improving systems for cyber security, particularly the intersection of generative AI, multi-agent learning, and game theory, to build safer and more resilient autonomous agents in adversarial environments.
Before starting my PhD, I worked as a master’s thesis research fellow at the Cyber-Defence Campus in Switzerland, training offensive penetration testing agents with multi-agent reinforcement learning (MARL). I also held a position as a research assistant in machine learning at the Institute for Computational Engineering (ICE) in Buchs, Switzerland, and gained several years of industry experience.
Research Interests
- Adversarial AI & AI Safety Developing robust AI systems that can withstand and adapt in adversarial settings, with a focus on cyber security.
- Multi-Agent & Equilibrium Learning Investigating cooperative and competitive interactions among autonomous agents using agentic, MARL, and game-theoretic frameworks.
- Generative AI for Security Leveraging generative models to explore both attack and defense in cyber environments.
- Reinforcement Learning Applying deep RL techniques to enable autonomous decision-making in complex, uncertain scenarios.
- Secure & Trustworthy Foundation Models Studying how to align and secure foundation models for deployment in critical infrastructure.
News
- Aug 2026Featured in MIT Technology Review (German edition, issue 6/2026) in the article Vergiftete Daten (Data Poisoning), discussing the threat of data poisoning to AI systems and my research on agentic, self-improving cybersecurity within the EU ELLIOT project.
- Jul 2026Our paper The Oracle's Gambit: A Game-Theoretic Framework for Responsible AI Release was accepted for an oral presentation at GameSec 2026, the 17th Conference on Game Theory and AI for Security.
- Feb 2026Our paper Enhancing Cyber Attack Autonomy Through Multi-Agent Reinforcement Learning (MARL) was accepted at CyCon 2026.
- Oct 2025Started as Teaching Assistant for Machine Learning in Cybersecurity (CISPA, Instructor: Prof. Dr. Mario Fritz), leading tutorial sessions and supporting exercises on ML security and AI applications.
- Oct 2025
- Aug 2025Attending the ELLIS Doctoral Symposium on Robust AI.
- Aug 2025
- Jun 2025Our paper Securing RAG: A Risk Assessment and Mitigation Framework was presented at the IEEE Swiss Conference on Data Science 2025.
- Jun 2025Gave a presentation on Multi-Agent Reinforcement Learning at the Cyber Alp Retreat Davos 2025.
- Jun 2025Poster presentation of our work Multi-Agent Reinforcement Learning in Cybersecurity: From Fundamentals to Applications at the ELSA General Assembly 2025.
- May 2025Started as an ELLIS PhD student.
Education
- PhD in Computer Science CISPA Helmholtz Center / ELLIS PhD Program, 2025 – present
- MSc in Engineering (Data Science) OST — Eastern Switzerland University of Applied Sciences. Thesis: Training of Offensive Penetration Testing Agents with Multi-Agent RL on Graphs (CYD Fellowship)
- BSc in Systems Engineering (Computer Science) OST — Eastern Switzerland University of Applied Sciences. Thesis: Anomaly Detection in Web Traffic using ML for Attack Recognition
Publications
A full list is available under Publications and on Google Scholar.
Contact
For collaboration opportunities, questions, or discussions related to my research, reach out by email or connect on LinkedIn.