English

A Note on Connectivity of Sublevel Sets in Deep Learning

Machine Learning 2021-01-22 v1 Machine Learning

Abstract

It is shown that for deep neural networks, a single wide layer of width N+1N+1 (NN being the number of training samples) suffices to prove the connectivity of sublevel sets of the training loss function. In the two-layer setting, the same property may not hold even if one has just one neuron less (i.e. width NN can lead to disconnected sublevel sets).

Keywords

Cite

@article{arxiv.2101.08576,
  title  = {A Note on Connectivity of Sublevel Sets in Deep Learning},
  author = {Quynh Nguyen},
  journal= {arXiv preprint arXiv:2101.08576},
  year   = {2021}
}
R2 v1 2026-06-23T22:23:08.804Z