061: Interpolation, Extrapolation and Linearisation (Prof. Yann LeCun, Dr. Randall Balestriero)
Machine Learning Street Talk (MLST)
00:00
Intro
This chapter explores the complexities of interpolation and extrapolation in machine learning, highlighting the inherent limitations of deep learning methods. It emphasizes how high-dimensional spaces blur the lines between true interpolation and extrapolation, particularly when dealing with complex datasets like images.
Transcript
Play full episode
