Empirical Study of the Benefits of Overparameterization in Learning Latent Variable ModelsDownload PDFOpen Website

2020 (modified: 19 Feb 2025)ICML 2020Readers: Everyone
Abstract: One of the most surprising and exciting discoveries in supervised learning was the benefit of overparameterization (i.e. training a very large model) to improving the optimization landscape of a pr...
0 Replies

Loading