Optimization for Machine Learning: SMO-MKL and Smoothing Strategies

Show simple item record

dc.contributor.author Vishwanathan, S. V. N. en_US
dc.date.accessioned 2011-04-27T20:14:25Z
dc.date.available 2011-04-27T20:14:25Z
dc.date.issued 2011-04-15
dc.identifier.uri http://hdl.handle.net/1853/38699
dc.description S V N Vishwanathan of Purdue University presented a lecture on April 15, 2011 from 2:00 pm - 3:00 pm in room 1447 of the Klaus Advanced Computing Building on the Georgia Tech campus. en_US
dc.description Runtime: 57:10 minutes en_US
dc.description.abstract Our objective is to train $p$-norm Multiple Kernel Learning (MKL) and, more generally, linear MKL regularised by the Bregman divergence, using the Sequential Minimal Optimization (SMO) algorithm. The SMO algorithm is simple, easy to implement and adapt, and efficiently scales to large problems. As a result, it has gained widespread acceptance and SVMs are routinely trained using SMO in diverse real world applications. Training using SMO has been a long standing goal in MKL for the very same reasons. Unfortunately, the standard MKL dual is not differentiable, and therefore can not be optimised using SMO style co-ordinate ascent. In this paper, we demonstrate that linear MKL regularised with the $p$-norm squared, or with certain Bregman divergences, can indeed be trained using SMO. The resulting algorithm retains both simplicity and efficiency and is significantly faster than the state-of-the-art specialised $p$-norm MKL solvers. We show that we can train on a hundred thousand kernels in less than fifteen minutes and on fifty thousand points in nearly an hour on a single core using standard hardware. en_US
dc.format.extent 57:10 minutes
dc.language.iso en_US en_US
dc.publisher Georgia Institute of Technology en_US
dc.relation.ispartofseries Computational Science and Engineering Seminar Series en_US
dc.subject Machine learning en_US
dc.subject Optimization en_US
dc.title Optimization for Machine Learning: SMO-MKL and Smoothing Strategies en_US
dc.type Lecture en_US
dc.type Video en_US
dc.contributor.corporatename Purdue University en_US

Files in this item

Files Size Format View Description
vishwanathan.mp4 150.8Mb MPEG-4 video View/ Open Download Video
vishwanathan_streaming.html 943bytes HTML View/ Open Streaming Video

This item appears in the following Collection(s)

Show simple item record