Zaid Harchaoui
INRIA Visual Recognition and Machine Learning Summer School
(http://www.harchaoui.eu/zaid/en/)
Monday 2nd April 2012
12 - 1pm
B10 Seminar Room, Basement,
Alexandra House, 17 Queen Square, London, WC1N 3AR
Learning with matrix gauge regularization penalty
Abstract:
We study learning problems with general sparsity-inducing matrix regularization penalties. We formulate the matrix regularizers as gauge functions, and, using their structure, we lift the optimization problem in a higher space where we pro- pose to apply a coordinate descent algorithm. Our framework allows to efficiently tackle difficult matrix-regularized objectives, e.g. with a trace-norm, or a group trace-norm, regularization penalty. We present experimental results on synthetic datasets and on real-world large-scale computer vision datasets. Our algorithm is competitive and often outperforms existing approaches on those problems.
Joint work with Miro Dudik and Jerome Malick