Weight-matrix probability densities in a toy autoencoder are evolved with the Fokker-Planck equation driven by the ADAM update, and the resulting output distributions roughly match training at epoch 5.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Fokker-Planck to Callan-Symanzik: evolution of weight matrices under training
Weight-matrix probability densities in a toy autoencoder are evolved with the Fokker-Planck equation driven by the ADAM update, and the resulting output distributions roughly match training at epoch 5.