Muideen Ogunniran

RESEARCH:

Muideen Ogunniran is a teacher in the Department of Mathematical Sciences at Osun State University who specializes in Computational Mathematics and Optimization. While working at the University of Chicago as an Eric and Wendy Schmidt AI in Science Faculty Fellow, Ogunniran will be researching numerical solutions for multi-dimensional partial differential equations (PDEs), including the Lindbald and Fokker Planck equations, using Rational Linear Multi-Step methods and the solutions’ machine-learning (ML) predictions. Solving PDEs provides physics-consistent solutions that serve as reliable training and validation data for ML models. By learning from these solutions, ML models can approximate the PDE’s behavior while respecting the physics-based constraints. Using ML models to compute PDEs provides a fast, real-time evaluative alternative to direct PDE solvers, which would be computationally expensive. For Ogunniran’s research, ML will help to optimize multi-dimensional PDEs using data from explicit numerical solution approaches by uncovering patterns in the equations, accelerating computation, and investigating new theoretical questions. Such fast, physics-constrained predictions can be applied to real-time control, engineering design optimization, and parameter estimation in systems governed by complex dynamics, including energy systems, fluid flows, and autonomous platforms.

BIO:

Muideen Ogunniran earned his PhD in Mathematics from the University of Ilorin, Nigeria, in 2019. His research worked to develop, analyze, and implement numerical schemes in integration singular and stiff problems on differential equations. By combining numerical schemes with ML, we gain accurate physical insights and efficient predictive capabilities beyond what either approach can currently provide alone. Practically, these schemes can be applied as fast surrogate models for real-time simulation, control, optimization, and inverse problems where repeated PDE solves are computationally prohibitive. Ogunniran’s Google Scholar Profile is accessible here.