Amirpasha Hedayat

prof_pic_2026.png

I am a Ph.D. candidate in Aerospace Engineering at the University of Michigan, advised by Prof. Karthik Duraisamy, and am also pursuing a Ph.D. in Scientific Computing through the Michigan Institute for Computational Discovery & Engineering (MICDE). I previously completed my M.A.Sc. in Mechanical Engineering at the University of British Columbia (2020-2022), where I worked with Prof. Carl Ollivier-Gooch, and received my B.Sc. in Aerospace Engineering from Amirkabir University of Technology (Tehran Polytechnic) (2015-2019).

My research focuses on scientific machine learning (SciML) and AI for science (AI4Science), with an emphasis on reduced-order modeling (model reduction), operator learning, and neural surrogates. My long-term goal is to develop fast and reliable physics-informed ML/AI models that enable near-real-time prediction, control, and decision-making for complex dynamical systems. Such capabilities are essential for emerging applications, including digital twins, design optimization, and intelligent simulators in science and engineering.

Questions, ideas, collaborations? Reach me at ahedayat@umich.edu.

news

Jul 15, 2026 Presented our work on In-Span Learning at the SIAM Annual Meeting (AN26) in Cleveland, OH.
Jul 03, 2026 Our preprint on In-Span Learning, adapting reduced-order models using their own predictions, is now available on arXiv.
Jun 02, 2026 Our paper on adaptive non-intrusive reduced-order models is now published in Structural and Multidisciplinary Optimization.