Deep linear neural networks in the proportional depth-width limit converge to a nontrivial mixture of Gaussians, with posterior output correlations that depend on the observed labels.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Proportional infinite-width infinite-depth limit for deep linear neural networks
Deep linear neural networks in the proportional depth-width limit converge to a nontrivial mixture of Gaussians, with posterior output correlations that depend on the observed labels.