A General Dimension for Approximately Learning Boolean Functions

  • Authors:
  • Johannes Köbler;Wolfgang Lindner

  • Affiliations:
  • -;-

  • Venue:
  • ALT '02 Proceedings of the 13th International Conference on Algorithmic Learning Theory
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

We extend the notion of general dimension, a combinatorial characterization of learning complexity for arbitrary query protocols, to encompass approximate learning. This immediately yields a characterization of the learning complexity in the statistical query model. As a further application, we consider approximate learning of DNF formulas and we derive close upper and lower bounds on the number of statistical queries needed. In particular, we show that with respect to the uniform distribution, and for any constant error parameter 驴 n variables and s terms) with tolerance 驴 = 驴(1/s) is n驴(log s).