### Abstract

A function defined on the Boolean hypercube is k-Fourier-sparse if it has at most k nonzero Fourier coefficients. For a function f: F^{n} _{2}→ ℝ and parameters k and d, we prove a strong upper bound on the number of k-Fourier-sparse Boolean functions that disagree with f on at most d inputs. Our bound implies that the number of uniform and independent random samples needed for learning the class of k-Fourier-sparse Boolean functions on n variables exactly is at most O(n · k log k). As an application, we prove an upper bound on the query complexity of testing Booleanity of Fourier-sparse functions. Our bound is tight up to a logarithmic factor and quadratically improves on a result due to Gur and Tamuz (Chicago J. Theor. Comput. Sci., 2013).

Original language | English (US) |
---|---|

Title of host publication | 30th Conference on Computational Complexity, CCC 2015 |

Publisher | Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing |

Pages | 58-71 |

Number of pages | 14 |

Volume | 33 |

ISBN (Print) | 9783939897811 |

DOIs | |

State | Published - Jun 1 2015 |

Event | 30th Conference on Computational Complexity, CCC 2015 - Portland, United States Duration: Jun 17 2015 → Jun 19 2015 |

### Other

Other | 30th Conference on Computational Complexity, CCC 2015 |
---|---|

Country | United States |

City | Portland |

Period | 6/17/15 → 6/19/15 |

### Fingerprint

### Keywords

- Fourier-sparse functions
- Learning theory
- List-decoding
- Property testing

### ASJC Scopus subject areas

- Software

### Cite this

*30th Conference on Computational Complexity, CCC 2015*(Vol. 33, pp. 58-71). Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing. https://doi.org/10.4230/LIPIcs.CCC.2015.58

**The list-decoding size of Fourier-sparse Boolean functions.** / Haviv, Ishay; Regev, Oded.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*30th Conference on Computational Complexity, CCC 2015.*vol. 33, Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing, pp. 58-71, 30th Conference on Computational Complexity, CCC 2015, Portland, United States, 6/17/15. https://doi.org/10.4230/LIPIcs.CCC.2015.58

}

TY - GEN

T1 - The list-decoding size of Fourier-sparse Boolean functions

AU - Haviv, Ishay

AU - Regev, Oded

PY - 2015/6/1

Y1 - 2015/6/1

N2 - A function defined on the Boolean hypercube is k-Fourier-sparse if it has at most k nonzero Fourier coefficients. For a function f: Fn 2→ ℝ and parameters k and d, we prove a strong upper bound on the number of k-Fourier-sparse Boolean functions that disagree with f on at most d inputs. Our bound implies that the number of uniform and independent random samples needed for learning the class of k-Fourier-sparse Boolean functions on n variables exactly is at most O(n · k log k). As an application, we prove an upper bound on the query complexity of testing Booleanity of Fourier-sparse functions. Our bound is tight up to a logarithmic factor and quadratically improves on a result due to Gur and Tamuz (Chicago J. Theor. Comput. Sci., 2013).

AB - A function defined on the Boolean hypercube is k-Fourier-sparse if it has at most k nonzero Fourier coefficients. For a function f: Fn 2→ ℝ and parameters k and d, we prove a strong upper bound on the number of k-Fourier-sparse Boolean functions that disagree with f on at most d inputs. Our bound implies that the number of uniform and independent random samples needed for learning the class of k-Fourier-sparse Boolean functions on n variables exactly is at most O(n · k log k). As an application, we prove an upper bound on the query complexity of testing Booleanity of Fourier-sparse functions. Our bound is tight up to a logarithmic factor and quadratically improves on a result due to Gur and Tamuz (Chicago J. Theor. Comput. Sci., 2013).

KW - Fourier-sparse functions

KW - Learning theory

KW - List-decoding

KW - Property testing

UR - http://www.scopus.com/inward/record.url?scp=84958249375&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958249375&partnerID=8YFLogxK

U2 - 10.4230/LIPIcs.CCC.2015.58

DO - 10.4230/LIPIcs.CCC.2015.58

M3 - Conference contribution

AN - SCOPUS:84958249375

SN - 9783939897811

VL - 33

SP - 58

EP - 71

BT - 30th Conference on Computational Complexity, CCC 2015

PB - Schloss Dagstuhl- Leibniz-Zentrum fur Informatik GmbH, Dagstuhl Publishing

ER -