Vision vs. Language
Ishan argues that computer vision presents greater challenges than linguistic intelligence, highlighting the fundamental nature of visual perception in intelligence. He discusses the differences in self-supervised learning approaches between vision and language, noting that while language benefits from a structured vocabulary, vision involves a vast combinatorial complexity that complicates predictions.In this clip
From this podcast
Lex Fridman Podcast
Ishan Misra: Self-Supervised Deep Learning in Computer Vision | Lex Fridman Podcast #206
Related Questions