"A Theoretical Explanation of Activation Sparsity through Flat Minima and ..."

Ze Peng et al. (2023)

Details and statistics

DOI: 10.48550/ARXIV.2309.03004

access: open

type: Informal or Other Publication

metadata version: 2023-12-05

a service of  Schloss Dagstuhl - Leibniz Center for Informatics