Sarra Alqahtani

About Sarra

Sarra Alqahtani is an Associate Professor of Computer Science at Wake Forest University whose research advances secure and explainable reinforcement learning, with a particular focus on Multi-Agent Reinforcement Learning (MARL). Her work develops methods for making autonomous AI agents more robust, trustworthy, and interpretable, especially in complex, high-stakes environments where multiple agents must learn, coordinate, and adapt under uncertainty. Her research investigates vulnerabilities, edge cases, and malicious behaviors that can emerge in neural-network-based decision-making systems. She designs software engineering techniques for debugging, monitoring, and fine-tuning reinforcement learning agents dynamically, reducing the need for complete retraining while improving system reliability and transparency. One application of this work involves heterogeneous swarms of autonomous drones that collaboratively navigate the Amazon rainforest to support secure monitoring of illegal gold mining operations. More broadly, her research contributes to the development of resilient AI systems for autonomous robotics, environmental monitoring, and safety-critical multi-agent coordination. In her lab, Alqahtani actively integrates undergraduate and graduate students into the research process, mentoring them in algorithm design, experimentation, publication, and the development of responsible AI systems.