Almost all monotone Boolean functions are polynomially learnable using membership queries

Authors:
Ilya Shmulevich;Aleksey D. Korshunov;Jaakko Astola
Affiliations:
Tampere Univ. of Technology, Tempere, Finland;Sobolev Institute of Mathematics, Novosibirsk, Russia;Tampere Univ. of Technology, Tempere, Finland
Venue:
Information Processing Letters
Year:
2001

Citing 5
Cited 5

Negative Results for Equivalence Queries

Machine Learning
Complexity of identification and dualization of positive Boolean functions

Information and Computation
Exact learning Boolean functions via the monotone theory

Information and Computation
Polynomial-Time Recognition of 2-Monotonic Positive Boolean Functions Given by an Oracle

SIAM Journal on Computing
Efficient Read-Restricted Monotone CNF/DNF Dualization by Learning with Membership Queries

Machine Learning

Hypergraph Transversal Computation and Related Problems in Logic and AI

JELIA '02 Proceedings of the European Conference on Logics in Artificial Intelligence
On algorithms for construction of all irreducible partial covers

Information Processing Letters
Computational aspects of monotone dualization: A brief survey

Discrete Applied Mathematics
On the complexity of monotone dualization and generating minimal hypergraph transversals

Discrete Applied Mathematics
On construction of the set of irreducible partial covers

SAGA'05 Proceedings of the Third international conference on StochasticAlgorithms: foundations and applications

Quantified Score

Hi-index	0.89

Visualization

Abstract

We consider exact learning or identification of monotone Boolean functions by only using membership queries. It is shown that almost all monotone Boolean functions are polynomially identifiable in the input number of variables as well as the output being the sum of the sizes of the CNF and DNF representations.