Quantum algorithms for the Goldreich-Levin learning problem
Quantum Physics
2020-01-03 v1
Abstract
The Goldreich-Levin algorithm was originally proposed for a cryptographic purpose and then applied to learning. The algorithm is to find some larger Walsh coefficients of an variable Boolean function. Roughly speaking, it takes a time to output the vectors with Walsh coefficients with probability at least . However, in this paper, a quantum algorithm for this problem is given with query complexity , which is independent of . Furthermore, the quantum algorithm is generalized to apply for an variable output Boolean function with query complexity .
Cite
@article{arxiv.2001.00014,
title = {Quantum algorithms for the Goldreich-Levin learning problem},
author = {Hongwei Li},
journal= {arXiv preprint arXiv:2001.00014},
year = {2020}
}
Comments
9 pages