collaborators

6 papers

math.PR2020

Fluctuations of extreme eigenvalues of sparse Erdős-Rényi graphs

Yukun He, Antti Knowles

We consider a class of sparse random matrices which includes the adjacency matrix of the Erdős-Rényi graph . We show that if $N^{\varepsilon} \leq Np \leq N^{1/3-…

cs.LG2018

Efficient Two-Step Adversarial Defense for Deep Neural Networks

Ting-Jui Chang, Yukun He, Peng Li

In recent years, deep neural networks have demonstrated outstanding performance in many machine learning tasks. However, researchers have discovered that these state-of-the-art mod…

math.PR2018

Local law and complete eigenvector delocalization for supercritical Erdős-Rényi graphs

Yukun He, Antti Knowles, Matteo Marcozzi

We prove a local law for the adjacency matrix of the Erdős-Rényi graph in the supercritical regime where has with high probability no isolated…

math.PR2018

Mesoscopic eigenvalue density correlations of Wigner matrices

Yukun He, Antti Knowles

We investigate to what extent the microscopic Wigner-Gaudin-Mehta-Dyson (WGMD) (or sine kernel) statistics of random matrix theory remain valid on mesoscopic scales. To that end, w…

math-ph2018

Diffusion Profile for Random Band Matrices: a Short Proof

Yukun He, Matteo Marcozzi

Let be a Hermitian random matrix whose entries are independent, centred random variables with variances , where $x, y \in (\mathbb Z/L\ma…

math.PR2018

Mesoscopic linear statistics of Wigner matrices of mixed symmetry class

Yukun He

We prove a central limit theorem for the mesoscopic linear statistics of Wigner matrices satisfying and , where…