Wednesday, February 24, 2021 11:00am to 12:00pm
About this Event
Title: "Complexity, Phase Transitions, and Inference"
Abstract: There is a deep analogy between Bayesian inference and statistical physics. Whenever we try to fit a model to noisy data, we can think about the “energy landscape" of possible models, and look for phase transitions where the ground truth suddenly gets lost in this landscape. I’ll use this framework to describe a phase transition in community detection in networks, where communities suddenly become hard or impossible to find. I will discuss why and how this detectability transition occurs, look at related spectral algorithms, and give a hint of similar phase transitions in other inference problems.
Please contact Molly Kruko (email@example.com) if interested in attending.
1 person is interested in this event