Samuel Marchal
Papers
1
Total Citations
7
H-Index
1
About
Samuel Marchal is a leading researcher at the intersection of cybersecurity and artificial intelligence, with a primary focus on adversarial machine learning and the robustness of deep reinforcement learning (DRL) systems. His most cited work, "Real-Time Adversarial Perturbations Against Deep Reinforcement Learning Policies: Attacks and Defenses" (2022, 7 citations), represents a pivotal contribution to understanding how DRL policies can be compromised in real-time environments. Marchal systematically characterizes vulnerabilities in DRL agents, demonstrating that even subtle, carefully crafted perturbations can cause catastrophic failures in autonomous decision-making—from robotics to game-playing AI. Beyond identifying these threats, he proposes novel defense mechanisms that balance security with computational efficiency, offering practical solutions for deploying DRL in safety-critical applications. His research has garnered attention for its rigorous experimental methodology and its implications for securing next-generation AI systems. Marchal’s work bridges a critical gap between theoretical adversarial examples and real-world deployment, making him a key voice in the ongoing dialogue about trustworthy AI. His contributions are particularly relevant for students and researchers exploring the fragility of learning-based control systems and the urgent need for resilient architectures.
Research Focus
Key Achievements
Top Papers
- 1