Google BERT Architecture Explained 1/3 - (BERT, Seq2Seq, Encoder Decoder)

Google BERT (Bidirectional Encoder Representations from Transformers) Machine Learning model for NLP has been a breakthrough. In this video series I am going to explain the architecture and help reducing time to understand the complex architecture.
Paper reference: Attention is all you need
Reference used in this part of Video:
ai.google/rese...
rajpurkar.gith...
google.github....
All References:
arxiv.org/pdf/...
github.com/hug...
mlexplained.com...
towardsdatasci...
towardsdatasci...
ai.google/rese...
rajpurkar.gith...
google.github....
lilianweng.git...
stats.stackexc...
Thanks to training partner: TechieGlobus : www.techieglobu...

[@adohreemas]

Finally someone made a good video on BERT. Hope more details will follow

[@carmelitamaia8535]

next time I recommend to remove the background music.it just creates noise in the presentation

[@SandeepBhutani]

Sure will take care.. Hope the content and delivery was relevant

[@ValeriaTiourina]

5:00 you say the language is German? It's Chinese :p ; thanks for the video!

[@SandeepBhutani]

Thanks.. I will learn both languages now 😉😊

[@ChauNguyen-kg4ok]

Japanese actually :)

[@epireve]

True, it’s Japanese.

[@mohdazam1404]

Its not german, Chinese :) , "Seher gut" . My comment is in german :)

[@SandeepBhutani]

I hope the Contents were useful to you

[@propropiro]

Confusing a bit and insecure ... many words but few Takeaways

[@SandeepBhutani]

Insecure means?

[@aqibfayyaz1619]

very well explained sir Awsome.

[@bharatbajoria]

Can you please explain building BERT for summarization task?

[@SandeepBhutani]

It is like clustering on BERT embeddings.. Did you face issues with links on Google?

[@DeepakChauhan-mn5jw]

You have an Indian flavor and a good way of explanation. Looking forward to your videos.

[@SandeepBhutani]

Thanks

[@SandeepBhutani]

Any specific content you are interested in?

[@DeepakChauhan-mn5jw]

@@SandeepBhutani I saw your playlist and most of the videos are on NLP. I think you can work on a playlist which can educate people in application of Machine learning and Deep learning for NLP problems. If you structure it well with an incremental storyline, I think that will have more value than university degrees/research work. Most of the people are scared away just by equations.

[@achintyaagarwal869]

Great Video.
I would also love to see the sequence of this video in which @Sandeep Bhutani, practically implements BERT in a Usecase/ project/example.
Keep up the good work.

[@SandeepBhutani]

Sure.. Will create a video on this

[@umadevig.r1759]

Hi Sandeep, I liked your explanation. Would you please make a video on how to implement this for Q&A task. Please explain the code. Thank you

[@SandeepBhutani]

Sure.. There is another video on QnA using allen NLP, in case you are interested

[@umadevig.r1759]

Thank you, I will go through it

[@shreyasvenugopalan1315]

Good overview. Thanks for the video.

[@Furcifer]

Not much technical aspects there, also try to improve your presentation style - it's very confusing. Thumbs up for the effort.

[@SandeepBhutani]

Thanks for the feedback.. What are u looking for? Code? Data values? Values at different layers?

[@RMCMC_AmitNikhade]

how can i feed multiple inputs to the seq2seq

[@SandeepBhutani]

The question is not clear.. Can you please elaborate your use case

[@RMCMC_AmitNikhade]

In a seq2seq model we just input a single input and get a single output... but is it possible to pass multiple inputs and get a single input

[@SandeepBhutani]

You can check caption generator.. There multiple inputs are being fed

[@kunalr_ai]

Farzi video

[@SandeepBhutani]

Thanks for your exper comment. It would be helpful if you can point out where and what went wrong in video.. Will take care in future