Abstract | ||
---|---|---|
In this note we disprove a conjecture of Kuzmin and Warmuth claiming that every family whose VC-dimension is at most d admits an unlabeled compression scheme to a sample of size at most d. We also study the unlabeled compression schemes of the joins of some families and conjecture that these give a larger gap between the VC-dimension and the size of the smallest unlabeled compression scheme for them. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1016/j.dam.2019.09.022 | Discrete Applied Mathematics |
Keywords | DocType | Volume |
Learning theory,Compression schemes,VC-dimension | Journal | 276 |
Issue | ISSN | Citations |
C | 0166-218X | 0 |
PageRank | References | Authors |
0.34 | 2 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dömötör Pálvölgyi | 1 | 202 | 29.14 |
Gábor Tardos | 2 | 1261 | 140.58 |