DeepAI AI Chat
Log In Sign Up

Sample compression schemes for balls in graphs

by   Jérémie Chalopin, et al.

One of the open problems in machine learning is whether any set-family of VC-dimension d admits a sample compression scheme of size O(d). In this paper, we study this problem for balls in graphs. For balls of arbitrary radius r, we design proper sample compression schemes of size 2 for trees, of size 3 for cycles, of size 4 for interval graphs, of size 6 for trees of cycles, and of size 22 for cube-free median graphs. For balls of a given radius, we design proper labeled sample compression schemes of size 2 for trees and of size 4 for interval graphs. We also design approximate sample compression schemes of size 2 for balls of δ-hyperbolic graphs.


page 1

page 2

page 3

page 4


Unlabeled Compression Schemes Exceeding the VC-dimension

In this note we disprove a conjecture of Kuzmin and Warmuth claiming tha...

Labeled sample compression schemes for complexes of oriented matroids

We show that the topes of a complex of oriented matroids (abbreviated CO...

A Labelled Sample Compression Scheme of Size at Most Quadratic in the VC Dimension

This paper presents a construction of a proper and stable labelled sampl...

Unlabelled Sample Compression Schemes for Intersection-Closed Classes and Extremal Classes

The sample compressibility of concept classes plays an important role in...

Agnostic Sample Compression for Linear Regression

We obtain the first positive results for bounded sample compression in t...

Breadth-First Rank/Select in Succinct Trees and Distance Oracles for Interval Graphs

We present the first succinct data structure for ordinal trees that supp...

Model Compression

With time, machine learning models have increased in their scope, functi...