Improved non-parametric sparse recovery with data matched penalties | IEEE Conference Publication | IEEE Xplore

Improved non-parametric sparse recovery with data matched penalties


Abstract:

This contribution studies the problem of learning sparse, nonparametric models from observations drawn from an arbitrary, unknown distribution. This specific problem lead...Show More

Abstract:

This contribution studies the problem of learning sparse, nonparametric models from observations drawn from an arbitrary, unknown distribution. This specific problem leads us to an algorithm extending techniques for Multiple Kernel Learning (MKL), functional ANOVA models and the Component Selection and Smoothing Operator (COSSO). The key element is to use a data-dependent regularization scheme adapting to the specific distribution underlying the data. We then present empirical evidence supporting the proposed learning algorithm.
Date of Conference: 14-16 June 2010
Date Added to IEEE Xplore: 14 October 2010
ISBN Information:

ISSN Information:

Conference Location: Elba, Italy

References

References is not available for this document.