Training deep neural networks from low-variance random weights drives a morphological instability that forms periodic channel structures in the weight distribution, independent of the data.
I., Keasar, C., Ben-Sasson, A
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Emergent weight morphologies in deep neural networks
Training deep neural networks from low-variance random weights drives a morphological instability that forms periodic channel structures in the weight distribution, independent of the data.