I work on optimal transport, optimization, sampling, and generative AI. My research develops theory and algorithms for learning and sampling from high-dimensional distributions. More specifically, I’m interested in how geometry, gradient flows, and low-dimensional structure can make sampling algorithms and generative models more efficient. I also work on problems in robust statistics and nonconvex optimization.

I am an Assistant Professor of Mathematics at Brandeis University. Previously, I was an Instructor of Applied Mathematics at MIT from 2018 to 2021. I received my PhD in Mathematics and MS in Statistics from the University of Minnesota in 2018.

Selected research

• Subspace Langevin Monte Carlo

  SIAM Journal on Mathematics of Data Science, 2026.
  arXiv

• Simplifying Optimal Transport through Schatten-p Regularization

  Transactions on Machine Learning Research, 2026.

• Global Convergence of Iteratively Reweighted Least Squares for Robust Subspace Recovery

  Preprint, 2025.

For questions, please email me at maunu@brandeis.edu.