Karnik Ram

I am an ELLIS PhD student at TU Munich and the University of Amsterdam, interested in physics-based deep learning and its applications in dynamical simulations. Read more about me, or take a look at my publications.

Karnik Ram

Recent Work

  • Learned Free-Energy Functionals from Pair-Correlation Matching for Dynamical Density Functional Theory
    Phys. Rev. E, 2025

    Karnik Ram, Jacobus Dijkman, René van Roij, Jan-Willem van de Meent, Bernd Ensing, Max Welling, Daniel Cremers

    arXiv
  • Trust-Region Noise Search for Black-Box Alignment of Diffusion and Flow Models
    ICLR 2026 ReALM-GEN Workshop

    Niklas Schweiger, Daniel Cremers*, Karnik Ram*

    Preprint
  • Evaluating Machine-Learned Inter-Atomic Potentials for a Practical Simulation Workflow
    ICLR 2026 Blogpost Track

    Richard Strunk, Daniel Cremers*, Karnik Ram*

    Preprint
All Publications