Chapter
Understanding Self-Supervised Learning in Natural Language Processing
44:02 - 49:24 (05:21)
Self-supervised learning works in natural language processing by predicting missing words in a test corpus, while uncertainty in prediction is much easier to represent than in image and video recognition. However, progress is being made in the latter fields.
Clips
In this podcast episode, the speaker explains self-supervised learning and how it differs from unsupervised learning.
44:02 - 46:52 (02:49)
Machine Learning
Summary
In this podcast episode, the speaker explains self-supervised learning and how it differs from unsupervised learning. They clarify that the algorithms used in self-supervised learning are the same as those used in supervised learning.
ChapterUnderstanding Self-Supervised Learning in Natural Language Processing
EpisodeYann LeCun: Deep Learning, Convolutional Neural Networks, and Self-Supervised Learning
PodcastLex Fridman Podcast
Self supervised learning works much better for natural language than it does for image recognition and video because it's easier to account for uncertainty in the prediction process.
46:53 - 49:24 (02:31)
Self Supervised Learning
Summary
Self supervised learning works much better for natural language than it does for image recognition and video because it's easier to account for uncertainty in the prediction process.