Профили пользователей по запросу "Jiaoyang Huang"
Jiaoyang HuangUniversity of Pennsylvania Подтвержден адрес электронной почты в домене wharton.upenn.edu Цитируется: 1445 |
Dynamics of deep neural networks and neural tangent hierarchy
The evolution of a deep neural network trained by the gradient descent in the
overparametrization regime can be described by its neural tangent kernel (NTK)\cite {jacot2018neural, …
overparametrization regime can be described by its neural tangent kernel (NTK)\cite {jacot2018neural, …
Edge universality of sparse random matrices
We consider the statistics of the extreme eigenvalues of sparse random matrices, a class of
random matrices that includes the normalized adjacency matrices of the Erd{\H o}sR{\'e}nyi …
random matrices that includes the normalized adjacency matrices of the Erd{\H o}sR{\'e}nyi …
Dyson Brownian motion for general and potential at the edge
A Adhikari, J Huang - Probability Theory and Related Fields, 2020 - Springer
In this paper, we compare the solutions of the Dyson Brownian motion for general $$\beta $$
β and potential V and the associated McKean–Vlasov equation near the edge. Under …
β and potential V and the associated McKean–Vlasov equation near the edge. Under …
How does information bottleneck help deep learning?
Numerous deep learning algorithms have been inspired by and understood via the notion of
information bottleneck, where unnecessary information is (often implicitly) minimized while …
information bottleneck, where unnecessary information is (often implicitly) minimized while …
Strong characterization for the Airy line ensemble
A Aggarwal, J Huang - arXiv preprint arXiv:2308.11908, 2023 - arxiv.org
In this paper we show that a Brownian Gibbsian line ensemble whose top curve approximates
a parabola must be given by the parabolic Airy line ensemble. More specifically, we prove …
a parabola must be given by the parabolic Airy line ensemble. More specifically, we prove …
Pearcey universality at cusps of polygonal lozenge tilings
We study uniformly random lozenge tilings of general simply connected polygons. Under a
technical assumption that is presumably generic with respect to polygon shapes, we show …
technical assumption that is presumably generic with respect to polygon shapes, we show …
Spectrum of random d‐regular graphs up to the edge
Consider the normalized adjacency matrices of random d‐regular graphs on N vertices with
fixed degree d ⩾ 3 $d\geqslant 3$ . We prove that, with probability 1 − N − 1 + ε $1-N^{-1+\…
fixed degree d ⩾ 3 $d\geqslant 3$ . We prove that, with probability 1 − N − 1 + ε $1-N^{-1+\…
[HTML][HTML] Bulk universality of sparse random matrices
We consider the adjacency matrix of the ensemble of Erdős-Rényi random graphs which
consists of graphs on N vertices in which each edge occurs independently with probability p. …
consists of graphs on N vertices in which each edge occurs independently with probability p. …
Edge rigidity of Dyson Brownian motion with general initial data
A Aggarwal, J Huang - Electronic Journal of Probability, 2024 - projecteuclid.org
In this paper, we study the edge behavior of Dyson Brownian motion with general β. Specifically,
we consider the scenario where the averaged initial density near the edge, on the scale …
we consider the scenario where the averaged initial density near the edge, on the scale …
Robustness implies generalization via data-dependent generalization bounds
This paper proves that robustness implies generalization via data-dependent generalization
bounds. As a result, robustness and generalization are shown to be connected closely in a …
bounds. As a result, robustness and generalization are shown to be connected closely in a …