This Thursday, 30.12 at 11.00 am we will have a FH seminar where Dr. Yohai Bar-Sinai from the Tel Aviv University will give a lecture titled "Calculating the entropy of physical systems with Machine Learning (and other stories)". See the details in the file attached.

Characterizing the entropy of a system is a crucial, and often computationally costly, step in
understanding its thermodynamics. We present a novel method, termed MICE (Machine-
learning Iterative Calculation of Entropy), for calculating the entropy of physical systems. Our
approach is to iteratively divide the system into smaller subsystems and estimate the mutual
entropy between each pair of halves. The estimation is performed with a recently proposed
machine learning algorithm which works with arbitrary network architectures that can fit the
structure and symmetries of the system at hand. We show that our method can calculate the
entropy of various systems, both thermal and athermal, with state-of-the-art accuracy, and
discuss promising future applications.
Time permitting, I will also briefly present a newly developed method to use Machine Learning
for the efficient solution of nonlinear PDEs. Our method learns approximate coarse-grid
representations, allowing us to integrate in time a collection of nonlinear equations at
resolutions 4-8x coarser than is possible with standard finite difference methods.