LESSWRONG
LW

775

Singular Learning Theory

Edited by DanielFilan last updated 20th Jun 2023

Singluar learning theory is a theory that applies algebraic geometry to statistical learning theory, developed by Sumio Watanabe. Reference textbooks are "the grey book", Algebraic Geometry and Statistical Learning Theory, and "the green book", Mathematical Theory of Bayesian Statistics.

1

1

Posts tagged Singular Learning Theory

84DSLT 0. Distilling Singular Learning Theory

2y

7

54DSLT 1. The RLCT Measures the Effective Dimension of Neural Networks

2y

10

31DSLT 3. Neural Networks are Singular

2y

5

31DSLT 2. Why Neural Networks obey Occam's Razor

2y

15

192Neural networks generalize because of this one weird trick

3y

34

92Singular learning theory: exercises

1y

6

188Announcing Timaeus

Jesse Hoogland, Daniel Murfet, Alexander Gietelink Oldenziel, Stan van Wingerden

2y

15

173Timaeus's First Four Months

Jesse Hoogland, Daniel Murfet, Stan van Wingerden, Alexander Gietelink Oldenziel

2y

6

94Investigating the learning coefficient of modular addition: hackathon project

Nina Panickssery, Dmitry Vaintrob

2y

5

90Growth and Form in a Toy Model of Superposition

Liam Carroll, Edmund Lau

2y

7

62Spooky action at a distance in the loss landscape

Jesse Hoogland, Filip Sondej

3y

4

37Gradient surfing: the hidden role of regularization

3y

9

34DSLT 4. Phase Transitions in Neural Networks

2y

3

194Towards Developmental Interpretability

Jesse Hoogland, Alexander Gietelink Oldenziel, Daniel Murfet, Stan van Wingerden

2y

10

113From SLT to AIT: NN generalisation out-of-distribution

10d

6

Load More (15/51)

Add Posts