ODSL Forum: Leopoldo Sarra, Renormalized Mutual Information for Artificial Scientific Discovery


Presenter: Leopoldo Sarra, Max Planck Institute for the Science of Light
Title: Renormalized Mutual Information for Artificial Scientific Discovery


We derive a well-defined "renormalized" version of mutual information that allows us to estimate the dependence between continuous random variables in the important case when one is deterministically dependent on the other. This quantity can be used 1) to estimate how much information a macroscopic quantity carries about a system and 2) to produce new low-dimensional effective descriptions of a high-dimensional system. This approach enables the discovery of collective variables in physical systems, or, more in general, to extract deterministic features via information maximization, thus adding to the toolbox of artificial scientific discovery.



As usual, the format will be a short presentation followed by plenty of discussion.

Please let all interested know about the Journal Club (get them to send their Email address).  We are looking forward to lively discussions.