Transformers – Why Self Attention Is Called Self ?


Why Self Attention Is Called Self ?

Table Of Contents:

  1. Why Self Attention Is Called Self ?

(1) Why Self Attention Is Called Self ?

  1. We have learnt the Attention concepts from the Luong Attention.
  2. In Luong attention mechanism, we calculate which word of the Encoder is more important in predicting the current time step output of the Decoder.
  3. To Do this we assign an attention score to each word of the Encoder and pass it as input to the Decoder.
  4. We put a SoftMax layer to normalize the attention score.
  5. The same operation mathematical we are performing in case of the Self Attention technique.
  6. Hence Self attention is also called as Attention technique.
  7. Why it is called Self because we are performing the attention technique to the same input sentence.
  8. In case of Luong attention the attention technique is performed in the two different input sentences.

(2) Why Is Self Attention Is Called Self ?

(3) Example Of Self Attention:

(4) Key Difference from Other Attention Types:

Leave a Reply

Your email address will not be published. Required fields are marked *