Tristan Bereau

Computational physicist and professor at the Institute for Theoretical Physics, Heidelberg University, whose work focuses on multiscale modeling and machine-learning methods for soft-condensed matter and biomolecules.

Split-Flows Make Mapping Entropy Computable for Molecular Coarse-Graining

Tristan Bereau presents Split-Flows, a flow-based method for connecting atomistic and coarse-grained molecular representations by adding explicit noise variables for the degrees of freedom lost under coarse-graining. The argument is that this augmentation turns a many-to-one mapping into a tractable coordinate transform, enabling both generative backmapping and computation of configuration-dependent mapping entropy. Bereau says the approach makes information loss measurable for complex molecular systems, though it depends on a differentiable bijective construction and still faces scaling costs.

Microsoft ResearchMay 26, 202617 min read