Scholar
Mohamad Louai Shehab
Google Scholar ID: DJvw8dUAAAAJ
University of Michigan
Citations & Impact
Citations
10
H-index
2
i10-index
0
Publications
4
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Learning Reward Machines from Partially Observed Policies' accepted at Transactions on Machine Learning Research (TMLR), 2025
- Paper 'Learning true objectives: Linear algebraic characterizations of identifiability in inverse reinforcement learning' accepted at Learning for Dynamics and Control Conference (L4DC)
- Passed Ph.D. Candidacy Qualifying Exam (CQE) in December 2022
- Passed thesis proposal exam in June 2025
- Presented research at multiple conferences including Purdue ICON, L4DC, and Midwest Workshop on Control and Game Theory
Background
- 5th-year Ph.D. student in the Robotics Department at the University of Michigan - Ann Arbor
- Research lies at the intersection of Reinforcement Learning, Inverse Reinforcement Learning, and Formal Methods
- Focuses on identifying and representing non-Markovian structure in sequential decision-making
- Develops algorithms that learn finite-state reward models (e.g., Reward Machines) from demonstrations or optimal policies, even under partial observability
- Aims to make reward learning more interpretable, identifiable, and generalizable by integrating tools from optimization, logic, and automata theory into maximum-entropy reinforcement learning frameworks
