Daniel Bourke
This month has been all about GPT3 and self-supervised learning.
GPT3 is the latest in language modelling from OpenAI and self-supervised learning seeks to understand the inherent underlying patterns of the data itself which can then be used for more specific tasks later. It can result in up to 1000x fewer labels being required and still achieve the same results as supervised learning.
Links:
ML Monthly July Edition – https://zerotomastery.io/blog/machine-learning-monthly-july-2020/
Learn Machine Learning – https://dbourke.link/mlcourse
Timestamps:
0:00 – Intro
0:29 – ML monthly July article
1:02 – My work during July
2:05 – All about GPT3
4:00 – GPT3 paper discussion
7:10 – What can GPT3 do?
9:39 – GPT3 vs. humans
11:20 – GPT3 learning resources
13:00 – Two trends I’m noticing in machine learning
17:45 – AutoML-Zero
19:30 – Facebook’s self-supervised image classification
21:50 – Learning resources
22:00 – Full-stack deep learning
23:10 – DeepMind’s deep learning course
24:37 – Self-supervised learning and computer vision
25:30 – Outro
Connect elsewhere:
Get email updates on my work – https://dbourke.link/newsletter
Support on Patreon – https://bit.ly/mrdbourkepatreon
Web – https://dbourke.link/web
Quora – https://dbourke.link/quora
Medium – https://dbourke.link/medium
Twitter – https://dbourke.link/twitter
LinkedIn – https://dbourke.link/linkedin