No article text was recorded for this version.
variance-preserving initialization
Version 1 of 132 papers34 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Invariance of Weight Distributions in Rectified MLPs
- Blog: Kolmogorov-Arnold Transformer
- Blog: Sinusoidal Initialization, Time for a New Start
- Blog: Classification and Prediction of Tibetan Medical Syndrome Based on the Improved BP Neural Network
- Blog: Principled Weight Initialisation for Input-Convex Neural Networks
- Blog: All you need is a good init Dmytro Mishkin
- Blog: Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
- Blog: Scaling Laws for the Principled Design, Initialization, and Preconditioning of ReLU Networks
- Blog: Implicit Regularization in Deep Matrix Factorization
- Blog: A scaling calculus for the design and initialization of ReLU networks
- Blog: Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
- Blog: Generic Perceptual Loss for Modeling Structured Output Dependencies
- Blog: How to Start Training: The Effect of Initialization and Architecture
- Blog: Adaptive Width Neural Networks
- Blog: Deep ReLU Networks Preserve Expected Length
- Blog: VI^3NR: Variance Informed Initialization for Implicit Neural Representations
- Blog: Self-Normalizing Neural Networks
- Blog: Initialization of Large Language Models via Reparameterization to Mitigate Loss Spikes
- Blog: On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias
- Blog: Understanding the difficulty of training deep feedforward neural networks
- Blog: Effectiveness of Self Normalizing Neural Networks for Text Classification
- Blog: Width Provably Matters in Optimization for Deep Linear Neural Networks
- Blog: Most Neural Networks Are Almost Learnable
- Blog: Deep Octonion Networks
- Blog: HyperPCM: Robust Task-Conditioned Modeling of Drug-Target Interactions
- Blog: Initializing neural networks for hierarchical multi-label text classification
- Blog: What Matters for On-Policy Deep Actor-Critic Methods? A Large-Scale Study
- Blog: Practical recommendations for gradient-based training of deep architectures
- Blog: Deep Video Deblurring: The Devil is in the Details
- Blog: Training wide residual networks for deployment using a single bit for each weight
- Blog: Simple initialization and parametrization of sinusoidal networks via their kernel bandwidth
- Blog: Recomposing the Reinforcement Learning Building Blocks with Hypernetworks
History & citation
Current · Permanent link
- This direction split from Principles of Variance Preserving Weight Initialization (72% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_variance_preserving_initialization_13fdef8b,
title = {variance-preserving initialization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/variance-preserving-initialization/rdr_13fdef8b26f640bda3c8f27509d5cb70},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_variance_preserving_initialization_13fdef8b_v1,
title = {variance-preserving initialization},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/variance-preserving-initialization/rdr_13fdef8b26f640bda3c8f27509d5cb70/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}