A Mean Field Theory of Batch Normalization
February 21, 2019 ยท Declared Dead ยท ๐ International Conference on Learning Representations
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Greg Yang, Jeffrey Pennington, Vinay Rao, Jascha Sohl-Dickstein, Samuel S. Schoenholz
arXiv ID
1902.08129
Category
cs.NE: Neural & Evolutionary
Cross-listed
cond-mat.dis-nn,
cs.LG,
math.DS
Citations
186
Venue
International Conference on Learning Representations
Last Checked
2 months ago
Abstract
We develop a mean field theory for batch normalization in fully-connected feedforward neural networks. In so doing, we provide a precise characterization of signal propagation and gradient backpropagation in wide batch-normalized networks at initialization. Our theory shows that gradient signals grow exponentially in depth and that these exploding gradients cannot be eliminated by tuning the initial weight variances or by adjusting the nonlinear activation function. Indeed, batch normalization itself is the cause of gradient explosion. As a result, vanilla batch-normalized networks without skip connections are not trainable at large depths for common initialization schemes, a prediction that we verify with a variety of empirical simulations. While gradient explosion cannot be eliminated, it can be reduced by tuning the network close to the linear regime, which improves the trainability of deep batch-normalized networks without residual connections. Finally, we investigate the learning dynamics of batch-normalized networks and observe that after a single step of optimization the networks achieve a relatively stable equilibrium in which gradients have dramatically smaller dynamic range. Our theory leverages Laplace, Fourier, and Gegenbauer transforms and we derive new identities that may be of independent interest.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Neural & Evolutionary
R.I.P.
๐ป
Ghosted
R.I.P.
๐ป
Ghosted
Progressive Growing of GANs for Improved Quality, Stability, and Variation
R.I.P.
๐ป
Ghosted
Learning both Weights and Connections for Efficient Neural Networks
R.I.P.
๐ป
Ghosted
LSTM: A Search Space Odyssey
R.I.P.
๐ป
Ghosted
A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
R.I.P.
๐ป
Ghosted
An Introduction to Convolutional Neural Networks
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Language Models are Few-Shot Learners
R.I.P.
๐ป
Ghosted
PyTorch: An Imperative Style, High-Performance Deep Learning Library
R.I.P.
๐ป
Ghosted
XGBoost: A Scalable Tree Boosting System
R.I.P.
๐ป
Ghosted