Completed
Implications: Depth Separation Result
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
A Function Space View of Overparameterized Neural Networks - Rebecca Willet, University of Chicago
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Overparameterized models in machine learning
- 3 An experiment
- 4 Training overparameterized neural nets
- 5 Approximation theory perspective
- 6 Infinite-width two-layer ReLU nets
- 7 Learning with norm-controlled infinite-width ReLU networks
- 8 From Two-layer ReLU Nets to Convex Nets
- 9 Intuition in 1D
- 10 Intuition in Higher Dimensions
- 11 The Radon Transform in 2D
- 12 Radon Transform as Line Detector
- 13 Key Derivation
- 14 Example
- 15 Implications: Comparison to Kernel Learning
- 16 Implications: Depth Separation Result
- 17 Open Questions
