Chapter
Limits of Self-Supervised Learning and Deep Learning
1:53:06 - 1:59:20 (06:14)
The limits of self-supervised learning and deep learning primarily involve the need for an interface to communicate with humans and the lack of guarantees in the system.
Clips
The limits of self-supervised learning are seen as needing an interface to communicate with the human, adding to the challenge of interacting meaningfully.
1:53:06 - 1:56:58 (03:51)
Self-Supervised Learning
Summary
The limits of self-supervised learning are seen as needing an interface to communicate with the human, adding to the challenge of interacting meaningfully. Another challenge is the lack of guarantees offered by machine learning and deep learning systems, particularly when it comes to self-supervised learning.
ChapterLimits of Self-Supervised Learning and Deep Learning
Episode#206 – Ishan Misra: Self-Supervised Deep Learning in Computer Vision
PodcastLex Fridman Podcast
Machine learning poses a challenge in determining what it means to be correct.
1:56:58 - 1:59:20 (02:22)
Machine Learning
Summary
Machine learning poses a challenge in determining what it means to be correct. There will always be scenarios in which a model will not perform optimally, leading to a certain level of nebulous correctness.