EasyDeepLearn

When do you use LayerNorm instead of BatchNorm?

medium

Answer

  • Use LayerNorm when batch size is small or varies (RNNs, transformers, sequence data).
  • LayerNorm computes statistics across features within a single example, so it doesn't depend on batch statistics.
  • That's why it's standard in transformers and language models where sequence lengths vary and batches can be tiny.
Check yourself — multiple choice
  • LayerNorm depends on batch size
  • LayerNorm normalizes across features within one example
  • BatchNorm is standard in transformers
  • LayerNorm has no learnable parameters

LayerNorm is per-example, making it robust to small/variable batches.

#normalization

Practise Deep Learning

214 interview questions in this topic.

Related questions