Current browse context:
cs.LG
Change to browse by:
References & Citations
Mathematical Physics
Title: Universal characteristics of deep neural network loss surfaces from random matrix theory
(Submitted on 17 May 2022 (v1), last revised 20 Jun 2022 (this version, v2))
Abstract: 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 related to local statistics to derive practical implications for deep neural networks based on a realistic model of their Hessians. In particular we derive universal aspects of outliers in the spectra of deep neural networks and demonstrate the important role of random matrix local laws in popular pre-conditioning gradient descent algorithms. We also present insights into deep neural network loss surfaces from quite general arguments based on tools from statistical physics and random matrix theory.
Submission history
From: Nicholas Baskerville [view email][v1] Tue, 17 May 2022 19:42:23 GMT (280kb,D)
[v2] Mon, 20 Jun 2022 21:34:04 GMT (280kb,D)
Link back to: arXiv, form interface, contact.