3.8k citations · 7.4k across the 34 of their papers we have counts for
Showing 2016Show all
2 papers · 1 filter
cs.AI2016★ 7 cited
Input Switched Affine Networks: An RNN Architecture Designed for Interpretability
Jakob N. Foerster, Justin Gilmer, Jan Chorowski +2
There exist many problem domains where the interpretability of neural network models is essential for deployment. Here we introduce a recurrent architecture composed of input-switc…
stat.ML2016★ 24 cited
Deep Information Propagation
Samuel S. Schoenholz, Justin Gilmer, Surya Ganguli +1
We study the behavior of untrained neural networks whose weights and biases are randomly distributed using mean field theory. We show the existence of depth scales that naturally l…