The sensitivity theorem


Date
Nov 16, 2021 3:00 PM - 4:00 PM
Location
[PST] UCLA Participating Analysis Seminar

he sensitivity theorem (former sensitivity conjecture) relates multiple ways to quantify the complexity, or lack of “smoothness”, of a boolean function f:{0,1}^n -> f : The minimum degree of a polynomial p(x):R^n -> R that extends f, the sensitivity s(f), and the block sensitivity bs(f). In 2019, H.Huang solved the conjecture with a remarkably short proof. I will give a self-contained explanation of this proof, and motivate the importance of the (former) conjecture by relating it to other measures of complexity for boolean functions.

Postdoctoral Researcher

Postroctoral researcher @ETHZ. My research interests include harmonic analysis (restriction, decoupling), Elliptic PDEs and deep learning theory.

Related