deep

Definition

Refers to stacking many layers so that representations compose from simple features to abstract ones. In practice "deep learning" is less a claim about layer counts than a label for the whole paradigm: end-to-end differentiable models that learn features instead of having them hand-designed. The adjective is now largely historical branding, retained because the field organized itself around it.

Explainers using this term

• Statistical Learning Theory — Why Does Learning Generalize?Understanding Deep Learning Requires Rethinking Generalization
• Surviving GPU Out-of-Memory — Every Cause, Every FixTraining Deep Nets with Sublinear Memory Cost
• Encoder or Decoder — The Fork in the Road Between BERT and GPTBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding