kmario23 / deep-learning-drizzle

Drench yourself in Deep Learning, Reinforcement Learning, Machine Learning, Computer Vision, and NLP by learning from these exciting lectures!!

Home Page:https://deep-learning-drizzle.github.io

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Transformers

kitagrawal opened this issue · comments

Maybe create a separate section on Transformers (in your to-do list). Recently, they have been getting a lot of attention.

@ankit--agrawal "attention", nice pun =)

:-P
There are some good blog posts on the topic but I will be particularly interested to know some good resources (lecture series) on this topic.

commented

Hey @ankit--agrawal ,
thanks for your suggestion! Since this is a specialized topic, let's maintain it in this thread, at least for now.

Here is a preliminary list of lectures:

Please feel free to suggest if I've overlooked any worthwhile lectures!!

Thank you so much for these. I will update the thread if I find something worthwhile. :)

Could you please make the Transformer list of lecture available on the main page?

Also how about adding content from the original attention papers:

  1. Neural Machine Translation by Jointly Learning to Align and Translate
    https://arxiv.org/pdf/1409.0473.pdf

  2. Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
    https://arxiv.org/pdf/1502.03044.pdf

Thanks in advance and let us know how we can lep further,
George

commented

Could you please make the Transformer list of lecture available on the main page?

This is a nice suggestion! I've been thinking of a neat way to add it on the main page.

Also how about adding content from the original attention papers:

  1. Neural Machine Translation by Jointly Learning to Align and Translate
    https://arxiv.org/pdf/1409.0473.pdf
  2. Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
    https://arxiv.org/pdf/1502.03044.pdf

I'm unsure about this since adding papers is not the goal of this repo!

Thanks in advance and let us know how we can lep further,
Contributions & suggestions are always welcome :)
George

I understand, there are plenty of lecture series on Transformers and blog posts that break down the original paper. I respect the framework you have chosen and I look forward your neat idea.