BERT (ang. Bidirectional Encoder Representations from Transformers), to nie głęboka sieć neuronowa! To transformator!
Polecamy każdemu zainteresowanemu tematem transfomratorów:
- Attention Is All You Need
- Transformer: A Novel Neural Network Architecture for Language Understanding
- Visualizing A Neural Machine Translation Model (Mechanics of Seq2seq Models With Attention)
- The Illustrated Transformer