Batch Normalization
Neural Networks
Normalizing layer inputs during training
What is Batch Normalization?
Normalizes inputs of each layer to have zero mean and unit variance, stabilizing and accelerating training.
Real-World Examples
- •Used in modern CNNs
- •Faster training convergence
- •Allows higher learning rates
When to Use This
Almost standard in deep networks for faster, stable training
Related Terms
Learn more about concepts related to Batch Normalization