Skip to content
lectern.tv
Deep Representation Learning & Approximation Theory →0/5
1

Lecture 01: Introduction to Deep Learning (6.7960)

Sara Beery · 1:00:51

2

Lecture 02: How to Train a Neural Net (6.7960)

Sara Beery · 1:19:34

3

Lecture 03: Approximation Theory (6.7960)

Jeremy Bernstein · 1:22:42

4

Lecture 1: SQL Optimization Background (15-799)

Andy Pavlo · 53:57

5

Lecture 2: History of Query Optimizers ft. IBM System R (15-799)

Andy Pavlo · 1:10:40

Lecture 03: Approximation Theory (6.7960)

Jeremy BernsteinMIT

Professor

1:22:42·2026·graduate level
Path step 3: Proves universal approximation properties and contrasts width versus depth expressivity.

This lecture investigates the approximation capabilities of deep neural networks, covering universal approximation theorems and Barron's theorem. It also discusses the impact of network depth on expressivity.

AI & MLDeep LearningApproximation TheoryBarron's TheoremNeural Network Expressivity

MIT 6.7960 Deep Learning, Fall 2024

Instructor: Jeremy Bernstein

View the complete course: https://ocw.mit.edu/courses/6-7960-deep-learning-fall-2024/

YouTube Playlist: https://www.youtube.com/playlist?list=PLUl4u3cNGP63URZnh5iqBzDTDYPUTQT-8

Previous · Step 2

Lecture 02: How to Train a Neural Net (6.7960)

Next · Step 4

Lecture 1: SQL Optimization Background (15-799)

Comments

Add a comment...Sign in