Member Seminar

Speaker: Changji Xu

Title: On the solution space of the Ising perceptron model

Abstract:  Consider the discrete cube $\{-1,1\}^N$ and a random collection of half spaces which includes each half space $H(x) := \{y \in \{-1,1\}^N: x \cdot y \geq \kappa \sqrt{N}\}$ for $x \in \{-1,1\}^N$ independently with probability $p$. The solution space is the intersection of these half spaces. In this talk, we will talk about its sharp threshold phenomenon, the frozen structure of the solution space, and the Gardner formula.