Learning Statistical Field Theories and Quantum Hamiltonians
Andrey Lokhov (Los Alamos National Laboratory)
Learning the interactions governing a physical system from observations is a central problem in statistical and quantum physics. Score matching offers an efficient approach without requiring the computation of partition functions. In this talk, I will present recent advances establishing rigorous guarantees on the amount of data needed for learning, together with extensions to statistical field theories and quantum Hamiltonians. I will show how these methods can recover interactions and reconstruct renormalization-group flows, and discuss generalizations that enable learning in lattice field theories with fermions and quantum systems.
Seminar of the Series MLP@P (Machine Learning Physics @ Plateau), joint with LISN and IPhT.
More info at: https://vale1925.wixsite.com/mlpatp