Reformer: The Efficient Transformer
This video explores the changes made in the Reformer to reduce memory bottlenecks and attend over long sequences. The following video and textbook chapter is extremely useful for getting a sense of Locality-Sensitive Hashing!
Mining Massive Datasets:
Video Lecture: https://www.youtube.com/watch?v=c6xK9WgRFhI
Book: http://infolab.stanford.edu/~ullman/mmds/ch3a.pdf
Reformer Paper Link: https://arxiv.org/pdf/2001.04451.pdf
Adaptive Attention Span: https://ai.facebook.com/blog/making-transformer-networks-simpler-and-more-efficient/
Sparse Transformers: https://openai.com/blog/sparse-transformer/
Видео Reformer: The Efficient Transformer канала Henry AI Labs
Mining Massive Datasets:
Video Lecture: https://www.youtube.com/watch?v=c6xK9WgRFhI
Book: http://infolab.stanford.edu/~ullman/mmds/ch3a.pdf
Reformer Paper Link: https://arxiv.org/pdf/2001.04451.pdf
Adaptive Attention Span: https://ai.facebook.com/blog/making-transformer-networks-simpler-and-more-efficient/
Sparse Transformers: https://openai.com/blog/sparse-transformer/
Видео Reformer: The Efficient Transformer канала Henry AI Labs
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Reformer: The Efficient Transformer
Is it easy to create your own Transformer? Everything you need to know about Transformers! || EB#42
CS480/680 Lecture 19: Attention and Transformer Networks
Transformer Neural Networks - EXPLAINED! (Attention is all you need)
NLP for Developers: Shrinking Transformers | Rasa
On the Measure of Intelligence (Introduction)
Generative Teaching Networks
MIT 6.S191 (2019): Recurrent Neural Networks![[Paper Review] Reformer: The Efficient Transformer](https://i.ytimg.com/vi/6ognBL6DEYM/default.jpg)
[Paper Review] Reformer: The Efficient Transformer
Funnel Transformer Explained!
Training GANs with Limited Data
Self-Supervised Learning
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
How does a Transformer work ?
GPT Explained!
Unsupervised Data Augmentation
Rasa Algorithm Whiteboard - Attention 1: Self Attention
Self-Training with Noisy Student (87.4% ImageNet Top-1 Accuracy!)
Self-Attention GAN