Completed
Theorem: How Much Space?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Example Memorization in Learning: Batch and Streaming - Differential Privacy for ML
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 What do we mean? "Memorizing training examples"
- 3 Empirical Example Memorization
- 4 Space, Information, and Deep Learning
- 5 Important preliminary: Shannon's mutual information
- 6 Theorem: Memorizing entire examples
- 7 Tasks: Mixtures of subpopulations
- 8 Tasks: Per-subpopulation distributions
- 9 Proof: Lower bounds via singletons
- 10 Experiments: Logistic regression and neural network
- 11 Setup: Learning from a stream of examples
- 12 Theorem: How Much Space?
- 13 Theorem: Example Memorization
- 14 Tasks: Space Lower Bounds for Natural Models
- 15 Proof: Structure and Overview
- 16 Proof: Requirements for distinguishing one bit
- 17 Main theorems and implications
- 18 Directions for future work
- 19 Memorize when you can't identify relevant information