A Nearly Optimal Variant of the Perceptron Algorithm for the Uniform Distribution on the Unit Sphere.
Marco SchmalhoferPublished in: COLT (2020)
Keyphrases
- uniform distribution
- unit sphere
- perceptron algorithm
- linear threshold functions
- pac learning
- boolean functions
- learning dnf
- mistake bound
- efficiently computable
- multi class
- target concept
- learning algorithm
- statistical queries
- linear threshold
- spherical harmonics
- worst case
- dynamic programming
- hyperplane
- term dnf
- computationally hard
- dnf formulas
- supervised learning
- small number
- support vector machine