Sign up to save your podcasts
Or
Share More powerful deep learning with transformers (Ep. 84)
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
More powerful deep learning with transformers (Ep. 84)
Some of the most powerful NLP models like BERT and GPT-2 have one thing in common: they all use the transformer architecture.
Such architecture is built on top of another important concept already known to the community: self-attention.
In this episode I explain what these mechanisms are, how they work and why they are so powerful.
Don't forget to subscribe to our Newsletter or join the discussion on our Discord server
References
- Attention is all you need
https://arxiv.org/abs/1706.03762
https://jalammar.github.io/illustrated-transformer
http://web.stanford.edu/class/cs224n/slides/cs224n-2019-lecture14-transformers.pdf
View all episodes
By Francesco Gadaleta
4.2
7272 ratings
Some of the most powerful NLP models like BERT and GPT-2 have one thing in common: they all use the transformer architecture.
Such architecture is built on top of another important concept already known to the community: self-attention.
In this episode I explain what these mechanisms are, how they work and why they are so powerful.
Don't forget to subscribe to our Newsletter or join the discussion on our Discord server
References
- Attention is all you need
https://arxiv.org/abs/1706.03762
https://jalammar.github.io/illustrated-transformer
http://web.stanford.edu/class/cs224n/slides/cs224n-2019-lecture14-transformers.pdf
More shows like Data Science at Home
View all
More or Less
889 Listeners
WSJ Tech News Briefing
1,646 Listeners
Software Engineering Daily
626 Listeners
Talk Python To Me
585 Listeners
BBC Inside Science
415 Listeners
Super Data Science: ML & AI Podcast with Jon Krohn
302 Listeners
FT Tech Tonic
99 Listeners
Worklife with Adam Grant
9,163 Listeners
Practical AI
211 Listeners
Last Week in AI
305 Listeners
Hard Fork
5,518 Listeners
This Day in AI Podcast
227 Listeners
The AI Daily Brief: Artificial Intelligence News and Analysis
610 Listeners
Unhedged
179 Listeners
The Last Invention
1,078 Listeners