Omar Saadi

Mohammed VI Polytechnic University (UM6P) in Benguerir, Morocco, is offering paid research internships in the field of optimization and reinforcement learning. The internships are hosted at the College of Computing and supervised by Assistant Professor Omar Saadi. The research topics include optimization methods for policy-gradient algorithms and analytical studies of acceleration in policy optimization, with a focus on advanced techniques such as proximal and trust-region methods (e.g., PPO). Interns will explore how update mechanisms and regularization strategies affect learning dynamics, and may engage in both algorithmic development and empirical evaluation on standard benchmarks. The projects are well-suited for students with a strong mathematical background and solid programming skills, particularly those interested in bridging optimization theory and reinforcement learning. Theoretical and analytical aspects will be emphasized, including mathematical modeling, numerical exploration, and conceptual analysis. These internships are research-oriented and may naturally lead to PhD opportunities in related areas. UM6P is a leading institution in Africa, dedicated to research and innovation, and aims to contribute to economic and human development on the continent. The university offers a unique partnership-based approach and is located in the heart of the Green City, Benguerir. The internships last 4–6 months and are paid, though specific stipend details are not provided. Applicants should apply via the provided links for each internship topic. The application process is through the university's career portal. For more information, candidates can refer to the supervisor's LinkedIn profile or the official university website.

PhD opportunity at Mohammed VI Polytechnic University (UM6P) , College of Computing, Benguerir, Morocco, in Optimization and Reinforcement Learning . The thesis topic is “Stability-Preserving Acceleration for Policy Optimization” , positioned at the intersection of optimization, reinforcement learning, and theoretical machine learning. The project focuses on efficient and reliable policy-optimization methods, with work spanning algorithmic development, mathematical analysis, and reproducible experiments. Ideal candidate profile: strong mathematical background, solid programming skills, interest in optimization and reinforcement learning, and motivation to balance theory with experiments. Supervisor/announcer: Omar Saadi, Assistant Professor at Mohammed VI Polytechnic University. The post is shared as a direct PhD recruitment call. Application: use the provided application link to apply. No deadline was stated in the post.