BERT: transfer learning for NLP

[@g111an]

Just binged the entire playlist, helped me understand the intuitions behind the math. I hope you make more videos :)

[@goelnikhils]

Thanks a lot Lennart. What a crisp and clear explanation of BERT.

[@rickyebay]

This is the best explanation on Transformer I have found in the web. Can you doing another set of video for T5 ?

[@jeremyyd1258]

Excellent video! Thank you!

[@JsaintUK]

Great video. Are the original word embeddings simple static embeddings? Where do they come from?

[@lennartsvensson7636]

They are "simple static embeddings". It is common to train them along with the other parameters.

[@JsaintUK]

@@lennartsvensson7636 Okay so they could be something such as Word2vec embeddings? These are then passed into the encoder where they are contextualised?

[@lennartsvensson7636]

@@JsaintUK They could be but more commonly they are trained along all the other network parameters.