Matt L. Sampson

prof_pic.jpg

Matt Sampson


Physics for AI


AI for Physics


PhD candidate at Princeton

Hey, I’m Matt, a PhD researcher at Princeton University working with Prof. Peter Melchior. I am studying ways to advance science and machine learning with a particular focus on modelling dynamical systems, representation learning, optimization, and world models. Previously I was at the Australian National University where I performed numerical experiments and helped develop a code to simulate cosmic ray propagation through a dynamically evolving turbulent plasma.

Long-term vision: To create models that deeply understand complex dynamical systems— enabling new discoveries and forming the foundations for increasingly general, physics-inspired intelligence.

news

May 13, 2025 Lucky to be one of 4 students from Princeton University nominated for the 2025 Google PhD Fellowship! The first nomination from the astrophysical sciences deptartment at Princeton

selected publications

  1. pathmin.png
    Path-minimizing latent ODEs for improved extrapolation and inference
    Matt L. Sampson and Peter Melchior
    Machine Learning: Science and Technology, Jun 2025
  2. cosmic.png
    Cosmic ray and plasma coupling for isothermal supersonic turbulence in the magnetized interstellar medium
    Matt L. Sampson, James R. Beattie, Romain Teyssier, and 5 more authors
    arXiv e-prints, Jun 2025
  3. score.png
    Score-matching neural networks for improved multi-band source separation
    Matt. L. Sampson, Peter. Melchior, Charlotte. Ward, and 1 more author
    Astronomy and Computing, Oct 2024
  4. hally.png
    Spotting Hallucinations in Inverse Problems with Data-Driven Priors
    Matt L. Sampson and Peter M. Melchior
    In ICML ML4Astrophysics Workshop (Oral), Jul 2023