DNF are teachable in the average case

Authors:
Homin K. Lee;Rocco A. Servedio;Andrew Wan
Affiliations:
Department of Computer Science, Columbia University, New York, USA 10027;Department of Computer Science, Columbia University, New York, USA 10027;Department of Computer Science, Columbia University, New York, USA 10027
Venue:
Machine Learning
Year:
2007

Citing 10
Cited 0

Interpolation and approximation of sparse multivariate polynomials over GF(2)

SIAM Journal on Computing
Learning binary relations and total orders

SIAM Journal on Computing
On the complexity of teaching

Journal of Computer and System Sciences
On specifying Boolean functions by labelled examples

Discrete Applied Mathematics
Witness sets for families of binary vectors

Journal of Combinatorial Theory Series A
Learning sparse multivariate polynomials over a field with queries and counterexamples

Journal of Computer and System Sciences
Simple Learning Algorithms for Decision Trees and Multivariate Polynomials

SIAM Journal on Computing
Learnability and Automatizability

FOCS '04 Proceedings of the 45th Annual IEEE Symposium on Foundations of Computer Science
Learning functions of k relevant variables

Journal of Computer and System Sciences - Special issue: STOC 2003
Teaching classes with high teaching dimension using few examples

COLT'05 Proceedings of the 18th annual conference on Learning Theory

Quantified Score

Hi-index	0.00

Visualization

Abstract

We study the average number of well-chosen labeled examples that are required for a helpful teacher to uniquely specify a target function within a concept class. This "average teaching dimension" has been studied in learning theory and combinatorics and is an attractive alternative to the "worst-case" teaching dimension of Goldman and Kearns which is exponential for many interesting concept classes. Recently Balbach showed that the classes of 1-decision lists and 2-term DNF each have linear average teaching dimension. As our main result, we extend Balbach's teaching result for 2-term DNF by showing that for any 1驴s驴2 驴(n), the well-studied concept classes of at-most-s-term DNF and at-most-s-term monotone DNF each have average teaching dimension O(ns). The proofs use detailed analyses of the combinatorial structure of "most" DNF formulas and monotone DNF formulas. We also establish asymptotic separations between the worst-case and average teaching dimension for various other interesting Boolean concept classes such as juntas and sparse GF 2 polynomials.