Karam Daaboul

Papers

Total Citations

H-Index

About

Karam Daaboul is a researcher specializing in safe reinforcement learning and constrained optimization for real-world control systems. His work addresses one of the most pressing challenges in applied machine learning: enabling autonomous agents to operate reliably and safely in environments where constraint violations carry meaningful consequences. His most notable contribution, "Safe Continuous Control with Constrained Model-Based Policy Optimization" (2021), tackles the fundamental tension between the asymptotic nature of classical reinforcement learning objectives and the strict safety requirements demanded by real-world deployments. By integrating model-based approaches with constrained policy optimization, Daaboul advances the field's ability to train agents that respect safety boundaries throughout the learning process — not merely at convergence. This work, which has garnered 10 citations since its publication, reflects growing recognition within the community of the importance of moving beyond unconstrained reward maximization toward principled safe exploration frameworks. For students and researchers working at the intersection of control theory, robotics, and machine learning, Daaboul's contributions offer a rigorous and practically motivated pathway toward deploying reinforcement learning in high-stakes, safety-critical applications.