English
Related papers

Related papers: Labeled compression schemes for extremal classes

200 papers

The sample compressibility of concept classes plays an important role in learning theory, as a sufficient condition for PAC learnability, and more recently as an avenue for robust generalisation in adaptive data analysis. Whether…

Machine Learning · Computer Science 2022-10-12 J. Hyam Rubinstein , Benjamin I. P. Rubinstein

This paper presents a construction of a proper and stable labelled sample compression scheme of size $O(\VCD^2)$ for any finite concept class, where $\VCD$ denotes the Vapnik-Chervonenkis Dimension. The construction is based on a well-known…

Machine Learning · Computer Science 2022-12-29 Farnam Mansouri , Sandra Zilles

Sample compression schemes were defined by Littlestone and Warmuth (1986) as an abstraction of the structure underlying many learning algorithms. Roughly speaking, a sample compression scheme of size $k$ means that given an arbitrary list…

Machine Learning · Computer Science 2015-04-15 Shay Moran , Amir Yehudayoff

We examine connections between combinatorial notions that arise in machine learning and topological notions in cubical/simplicial geometry. These connections enable to export results from geometry to machine learning. Our first main result…

Discrete Mathematics · Computer Science 2022-03-03 Jérémie Chalopin , Victor Chepoi , Shay Moran , Manfred K. Warmuth

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…

Combinatorics · Mathematics 2021-10-15 Dömötör Pálvölgyi , Gábor Tardos

This work studies embedding of arbitrary VC classes in well-behaved VC classes, focusing particularly on extremal classes. Our main result expresses an impossibility: such embeddings necessarily require a significant increase in dimension.…

Discrete Mathematics · Computer Science 2024-05-28 Zachary Chase , Bogdan Chornomaz , Steve Hanneke , Shay Moran , Amir Yehudayoff

A hypothesis class admits a sample compression scheme, if for every sample labeled by a hypothesis from the class, it is possible to retain only a small subsample, using which the labels on the entire sample can be inferred. The size of the…

Machine Learning · Computer Science 2023-09-22 Chirag Pabbaraju

Resolving a conjecture of Littlestone and Warmuth, we show that any concept class of VC-dimension $d$ has a sample compression scheme of size $d$.

Machine Learning · Computer Science 2022-01-14 Zachary Chase

We show that the topes of a complex of oriented matroids (abbreviated COM) of VC-dimension $d$ admit a proper labeled sample compression scheme of size $d$. This considerably extends results of Moran and Warmuth on ample classes, of…

Combinatorics · Mathematics 2023-04-21 Victor Chepoi , Kolja Knauer , Manon Philibert

The Sample Compression Conjecture of Littlestone & Warmuth has remained unsolved for over two decades. This paper presents a systematic geometric investigation of the compression of finite maximum concept classes. Simple arrangements of…

Machine Learning · Computer Science 2014-02-04 Benjamin I. P. Rubinstein , J. Hyam Rubinstein

Term Coding asks: given a finite system of term identities $\Gamma$ in $v$ variables, how large can its solution set be on an $n$--element alphabet, when we are free to choose the interpretations of the function symbols? This turns familiar…

Information Theory · Computer Science 2026-01-26 Søren Riis

The Sauer-Shelah-Perles Lemma is a cornerstone of combinatorics and learning theory, bounding the size of a binary hypothesis class in terms of its Vapnik-Chervonenkis (VC) dimension. For classes of functions over a $k$-ary alphabet, namely…

Machine Learning · Computer Science 2026-04-15 Steve Hanneke , Qinglin Meng , Shay Moran , Amirreza Shaeiri

The Vapnik-Chervonenkis dimension is a combinatorial parameter that reflects the "complexity" of a set of sets (a.k.a. concept classes). It has been introduced by Vapnik and Chervonenkis in their seminal 1971 paper and has since found many…

Machine Learning · Computer Science 2015-07-21 Shai Ben-David

In the realm of machine learning theory, to prevent unnatural coding schemes between teacher and learner, No-Clash Teaching Dimension was introduced as provably optimal complexity measure for collusion-free teaching. However, whether…

Information Theory · Computer Science 2026-04-02 Jiahua Liu , Benchong Li

It was proved in 1998 by Ben-David and Litman that a concept space has a sample compression scheme of size d if and only if every finite subspace has a sample compression scheme of size d. In the compactness theorem, measurability of the…

Machine Learning · Statistics 2015-03-20 Damjan Kalajdzievski

Statistical learning theory chiefly studies restricted hypothesis classes, particularly those with finite Vapnik-Chervonenkis (VC) dimension. The fundamental quantity of interest is the sample complexity: the number of samples required to…

Machine Learning · Computer Science 2008-07-10 David Soloveichik

In this work we study the quantitative relation between VC-dimension and two other basic parameters related to learning and teaching. Namely, the quality of sample compression schemes and of teaching sets for classes of low VC-dimension.…

Machine Learning · Computer Science 2016-11-28 Shay Moran , Amir Shpilka , Avi Wigderson , Amir Yehudayoff

Set systems of finite VC dimension are frequently used in applications relating to machine learning theory and statistics. Two simple types of VC classes which have been widely studied are the maximum classes (those which are extremal with…

Probability · Mathematics 2013-09-11 Hunter Johnson

Vapnik-Chervonenkis (VC) dimension is a fundamental measure of the generalization capacity of learning algorithms. However, apart from a few special cases, it is hard or impossible to calculate analytically. Vapnik et al. [10] proposed a…

Machine Learning · Statistics 2011-11-16 Daniel J. McDonald , Cosma Rohilla Shalizi , Mark Schervish

In Statistical Learning, the Vapnik-Chervonenkis (VC) dimension is an important combinatorial property of classifiers. To our knowledge, no theoretical results yet exist for the VC dimension of edited nearest-neighbour (1NN) classifiers…

Machine Learning · Computer Science 2019-02-08 Iain A. D. Gunn , Ludmila I. Kuncheva
‹ Prev 1 2 3 10 Next ›