HSE SLT, lecture 13: proof of lazy training in wide networks

Name: HSE SLT, lecture 13: proof of lazy training in wide networks
Uploaded: 2025-12-09T16:11:28+03:00
Duration: 1 h 20 min 26 s
Channel: BrunosTCSlectures
Description: - Proof that parameter updates in wide networks have small norm - Neural tangent kernel and its min-eigenvalue - Solving linear systems with gradient descent, condition number, link to double descent - Label dependent bound on the distance and optional tasks (see chapter 18 of the notes) Course webs

12+

3 просмотра

9 месяцев назад

- Proof that parameter updates in wide networks have small norm - Neural tangent kernel and its min-eigenvalue - Solving linear systems with gradient descent, condition number, link to double descent - Label dependent bound on the distance and optional tasks (see chapter 18 of the notes) Course website http://wiki.cs.hse.ru/Statistical_learning_theory_2025