AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Critical Initialization of Wide and Deep Neural Networks using Partial Jacobians: General Theory and Applications

Darshil Doshi, Tianyu He, Andrey Gromov

NeurIPS 2023 · Architectures and efficiency › RNNs and normalization · approach: Theory of depth and width · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add residual connections with strength mu=1 to LayerNorm MLP (ablation)deep MLP (L=50, N=500) · FashionMNISTtraining accuracyImproved
use erf activation with mu=1 residual connections without LayerNorm (subcritical case) (ablation)deep MLP (L=50, N=500) · FashionMNISTtraining accuracyImproved
decrease sigma_w in erf MLP with mu=1 residual connections (ablation)deep MLP with erf · FashionMNISTtraining accuracyImproved
increase residual strength mu (ablation)ResNet110 (LayerNorm) · training accuracy vs sigma_w^2training accuracyImproved
increase mu monotonically at (sigma_w^2, sigma_b^2) = (2,0) (ablation)ResNet110 (LayerNorm) · training accuracy vs mutraining accuracyImproved
set mu=0 with small sigma_w^2 (ablation)ResNet110 (LayerNorm) · training accuracytraining accuracyHurt
increase mu from 0.5 to 1 (ablation)MLP-Mixer (L=100 blocks) · CIFAR-10 training curvestraining accuracy: 0.5 -> 1Improved
use mu=1 residual connection with BatchNormdeep network · trainabilitytrainabilityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy