ML Anthology

PAC-Bayes Learning of Conjunctions and Classification of Gene-Expression Data

Mario Marchand, Mohak Shah
NeurIPS 2004 pp. 881-888
/neurips/2004/marchand2004neurips-pacbayes/

Abstract

We propose a “soft greedy” learning algorithm for building small conjunctions of simple threshold functions, called rays, defined on single real-valued attributes. We also propose a PAC-Bayes risk bound which is minimized for classifiers achieving a non-trivial tradeoff between sparsity (the number of rays used) and the mag- nitude of the separating margin of each ray. Finally, we test the soft greedy algorithm on four DNA micro-array data sets.

PDF NeurIPS Semantic Scholar

Cite

Text

Marchand and Shah. "PAC-Bayes Learning of Conjunctions and Classification of Gene-Expression Data." Neural Information Processing Systems, 2004.

Markdown

[Marchand and Shah. "PAC-Bayes Learning of Conjunctions and Classification of Gene-Expression Data." Neural Information Processing Systems, 2004.](https://mlanthology.org/neurips/2004/marchand2004neurips-pacbayes/)

BibTeX

@inproceedings{marchand2004neurips-pacbayes,
  title     = {{PAC-Bayes Learning of Conjunctions and Classification of Gene-Expression Data}},
  author    = {Marchand, Mario and Shah, Mohak},
  booktitle = {Neural Information Processing Systems},
  year      = {2004},
  pages     = {881-888},
  url       = {https://mlanthology.org/neurips/2004/marchand2004neurips-pacbayes/}
}