Rasa Algorithm Whiteboard - Attention 3: Multi Head Attention
This is the third video on attention mechanisms. In the previous video we introduced keys, queries and values and in this video we're introducing the concept of multiple heads.
The colab notebook that contains the interactive visualisation can be found here:
https://colab.research.google.com/github/tensorflow/tensor2tensor/blob/master/tensor2tensor/notebooks/hello_t2t.ipynb#scrollTo=T7UJzFf6fmhp
We're going at it step by step, but if you're interested in immediately reading all about it in full detail then we might recommend these online documents:
- http://www.peterbloem.nl/blog/transformers
- http://jalammar.github.io/illustrated-transformer/
- http://d2l.ai/chapter_attention-mechanisms/attention.html
The general github repo for this playlist can be found here: https://github.com/RasaHQ/algorithm-whiteboard-resources.
Видео Rasa Algorithm Whiteboard - Attention 3: Multi Head Attention канала Rasa
The colab notebook that contains the interactive visualisation can be found here:
https://colab.research.google.com/github/tensorflow/tensor2tensor/blob/master/tensor2tensor/notebooks/hello_t2t.ipynb#scrollTo=T7UJzFf6fmhp
We're going at it step by step, but if you're interested in immediately reading all about it in full detail then we might recommend these online documents:
- http://www.peterbloem.nl/blog/transformers
- http://jalammar.github.io/illustrated-transformer/
- http://d2l.ai/chapter_attention-mechanisms/attention.html
The general github repo for this playlist can be found here: https://github.com/RasaHQ/algorithm-whiteboard-resources.
Видео Rasa Algorithm Whiteboard - Attention 3: Multi Head Attention канала Rasa
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Rasa Algorithm Whiteboard: Attention 4 - TransformersTransformer Neural Networks - EXPLAINED! (Attention is all you need)Rasa Algorithm Whiteboard - Attention 1: Self AttentionBERT Research - Ep. 6 - Inner Workings III - Multi-Headed AttentionRasa Algorithm Whiteboard - Response Selection: ImplementationAttention is all you need; Attentional Neural Network Models | Łukasz Kaiser | MasterclassRasa Algorithm Whiteboard - Subword EmbeddingsHopfield Networks is All You Need (Paper Explained)Rasa Algorithm Whiteboard - TED PolicyPytorch Transformers from Scratch (Attention is all you need)Why Does Batch Norm Work? (C2W3L06)NLP with Neural Networks & TransformersRasa Algorithm Whiteboard - Response SelectionAttention in Neural NetworksRasa Algorithm Whiteboard: Bias in Word EmbeddingsBERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingShould you switch from BERT to ALBERT?Rasa Algorithm Whiteboard - TED in PracticeRasa Algorithm Whiteboard - Why Debiasing is Hard