Generalized spike-and-slab priors for Bayesian group feature selection using expectation propagation

@article{HernndezLobato2013GeneralizedSP,
  title={Generalized spike-and-slab priors for Bayesian group feature selection using expectation propagation},
  author={Daniel Hern{\'a}ndez-Lobato and Jos{\'e} Miguel Hern{\'a}ndez-Lobato and Pierre Dupont},
  journal={Journal of Machine Learning Research},
  year={2013},
  volume={14},
  pages={1891-1945}
}
We describe a Bayesian method for group feature selection in li ear regression problems. The method is based on a generalized version of the standard spik e and slab prior distribution which is often used for individual feature selection. Exact Bayes ian inference under the prior considered is infeasible for typical regression problems. However, ap… CONTINUE READING