English

Quantum Algorithms for Learning and Testing Juntas

Quantum Physics 2007-10-16 v1 Machine Learning

Abstract

In this article we develop quantum algorithms for learning and testing juntas, i.e. Boolean functions which depend only on an unknown set of k out of n input variables. Our aim is to develop efficient algorithms: - whose sample complexity has no dependence on n, the dimension of the domain the Boolean functions are defined over; - with no access to any classical or quantum membership ("black-box") queries. Instead, our algorithms use only classical examples generated uniformly at random and fixed quantum superpositions of such classical examples; - which require only a few quantum examples but possibly many classical random examples (which are considered quite "cheap" relative to quantum examples). Our quantum algorithms are based on a subroutine FS which enables sampling according to the Fourier spectrum of f; the FS subroutine was used in earlier work of Bshouty and Jackson on quantum learning. Our results are as follows: - We give an algorithm for testing k-juntas to accuracy ϵ\epsilon that uses O(k/ϵ)O(k/\epsilon) quantum examples. This improves on the number of examples used by the best known classical algorithm. - We establish the following lower bound: any FS-based k-junta testing algorithm requires Ω(k)\Omega(\sqrt{k}) queries. - We give an algorithm for learning kk-juntas to accuracy ϵ\epsilon that uses O(ϵ1klogk)O(\epsilon^{-1} k\log k) quantum examples and O(2klog(1/ϵ))O(2^k \log(1/\epsilon)) random examples. We show that this learning algorithms is close to optimal by giving a related lower bound.

Keywords

Cite

@article{arxiv.0707.3479,
  title  = {Quantum Algorithms for Learning and Testing Juntas},
  author = {Alp Atici and Rocco A. Servedio},
  journal= {arXiv preprint arXiv:0707.3479},
  year   = {2007}
}

Comments

15 pages, 1 figure. Uses synttree package. To appear in Quantum Information Processing

R2 v1 2026-06-21T09:01:06.385Z