Math Physics Seminar
“Machine Learning Analysis of Ising Worms” by Professor Yannick Meurice and Mr. Samuel Foreman, Department of Physics & Astronomy, The University of Iowa
Abstract: We discuss the sampling of high temperature configurations for the two-dimensional Ising model (“worms”). We show that worm averages can be used to calculate the thermodynamic energy and the known logarithmic divergence of the specific heat at the critical temperature. We show numerical evidence supporting the conjecture that the leading eigenvalue of the PCA, a method commonly used to deal with images in machine learning, has a logarithmic divergence at the critical temperature. We discuss the correspondence between the two approaches under coarse graining procedures.