Peyman Mohajerin Esfahani Associate Professor Mechanical & Industrial Engineering (Operations Research) Electrical & Computer Engineering (Systems Control, cross-appointment) University of Toronto Office: MC321 Phone: +1 416 978 0164 Email: P.MohajerinEsfahani@utoronto.ca

Research Interests

My research interests revolve around theoretical and practical aspects of decision-making problems in uncertain and dynamic environments, with applications in engineering systems and management. In my research, I use tools from Control Theory, Optimization, Machine Learning, and Information Theory.

News

• July 2026: New papers on a mathematical model for LLM training dynamics under noisy reward (ICML 2026)

  • Rate or Fate? RLV\(^\varepsilon\)R: Reinforcement Learning with Verifiable Noisy Rewards, [Blog] (Spotlight Paper)

  • G\(^2\)RPO: Geometric GRPO Escaping LLM’s Reasoning Rut to Break Accuracy–Entropy Trade-off

• May 2026: New paper on generalization bounds in inverse optimization
  \(\Rightarrow\) Tight Generalization Bounds for Noiseless Inverse Optimization

• May 2026: Invited talk at Optimal Transport + Optimization Workshop, Switzerland

• April 2026: New paper on maching in optimal transport via coupling dynamics (SIAM Journal on Scientific Computing)
  \(\Rightarrow\) Optimal Transportation by Orthogonal Coupling Dynamics

• December 2025: Survey paper on fault diagnosis in dynamical systems with a geometric perspective
  \(\Rightarrow\) Fault Diagnosis in Dynamical Systems: Geometric Interpretation and Tractable Algorithms

• December 2025: Survey paper on Control and RL with a focus on algorithmic aspects
  \(\Rightarrow\) Control and Reinforcement Learning through the Lens of Optimization: An Algorithmic Perspective