English

Tight Bounds on Proper Equivalence Query Learning of DNF

Machine Learning 2011-11-07 v1 Computational Complexity

Abstract

We prove a new structural lemma for partial Boolean functions ff, which we call the seed lemma for DNF. Using the lemma, we give the first subexponential algorithm for proper learning of DNF in Angluin's Equivalence Query (EQ) model. The algorithm has time and query complexity 2(O~n)2^{(\tilde{O}{\sqrt{n}})}, which is optimal. We also give a new result on certificates for DNF-size, a simple algorithm for properly PAC-learning DNF, and new results on EQ-learning logn\log n-term DNF and decision trees.

Keywords

Cite

@article{arxiv.1111.1124,
  title  = {Tight Bounds on Proper Equivalence Query Learning of DNF},
  author = {Lisa Hellerstein and Devorah Kletenik and Linda Sellie and Rocco Servedio},
  journal= {arXiv preprint arXiv:1111.1124},
  year   = {2011}
}