186 citations · 563 across the 31 of their papers we have counts for
19 papers · 1 filter
Universal characteristics of deep neural network loss surfaces from random matrix theory
Nicholas P Baskerville, Jonathan P Keating, Francesco Mezzadri +2
This paper considers several aspects of random matrix universality in deep neural networks. Motivated by recent experimental work, we use universal properties of random matrices re…
On the moments of characteristic polynomials
Bhargavi Jonnadula, Jon Keating, Francesco Mezzadri
We examine the asymptotics of the moments of characteristic polynomials of matrices drawn from the Hermitian ensembles of Random Matrix Theory, in the limit as $N\to\in…
A spin-glass model for the loss surfaces of generative adversarial networks
Nicholas P Baskerville, Jonathan P Keating, Francesco Mezzadri +1
We present a novel mathematical model that seeks to capture the key design feature of generative adversarial networks (GANs). Our model consists of two interacting spin glasses, an…
Symmetric Function Theory and Unitary Invariant Ensembles
Bhargavi Jonnadula, Jonathan P. Keating, Francesco Mezzadri
Representation theory and the theory of symmetric functions have played a central role in Random Matrix Theory in the computation of quantities such as joint moments of traces and…
Entanglement entropy of two disjoint intervals separated by one spin in a chain of free fermion
L. Brightmore, G. P. Geher, A. R. Its +4
We calculate the entanglement entropy of a non-contiguous subsystem of a chain of free fermions. The starting point is a formula suggested by Jin and Korepin, \texttt{arXiv:1104.10…
Free fermions and the classical compact groups
Fabio Deelan Cunden, Francesco Mezzadri, Neil O'Connell
There is a close connection between the ground state of non-interacting fermions in a box with classical (absorbing, reflecting, and periodic) boundary conditions and the eigenvalu…