Attribute-Efficient and Non-Adaptive Learning of Parities and DNF Expressions

Abstract

We consider the problems of attribute-efficient PAC learning of two well-studied concept classes: parity functions and DNF expressions over 0,1n. We show that attribute-efficient learning of parities with respect to the uniform distribution is equivalent to decoding high-rate random linear codes from low number of errors, a long-standing open problem in coding theory. This is the first evidence that attribute-efficient learning of a natural PAC learnable concept class can be computationally hard.

Cite

Text

Feldman. "Attribute-Efficient and Non-Adaptive Learning of Parities and DNF Expressions." Journal of Machine Learning Research, 2007.

Markdown

[Feldman. "Attribute-Efficient and Non-Adaptive Learning of Parities and DNF Expressions." Journal of Machine Learning Research, 2007.](https://mlanthology.org/jmlr/2007/feldman2007jmlr-attributeefficient/)

BibTeX

@article{feldman2007jmlr-attributeefficient,
  title     = {{Attribute-Efficient and Non-Adaptive Learning of Parities and DNF Expressions}},
  author    = {Feldman, Vitaly},
  journal   = {Journal of Machine Learning Research},
  year      = {2007},
  pages     = {1431-1460},
  volume    = {8},
  url       = {https://mlanthology.org/jmlr/2007/feldman2007jmlr-attributeefficient/}
}