On Tue, Oct 11, 2011 at 10:43 PM, Olivier Grisel <[email protected]> wrote: > I think people here (e.g. @pprett) might be interested in the > following new release of Léon Bottou's influential project: > > http://leon.bottou.org/projects/sgd > > It did not know about Averaged SGD. Will have to read the cited references. > > I wonder if those results are transposable on online clustering and / > or online matrix factorization.
I think it should be, because those results look similar to arguments made by e.g. Cappe and Moulines in a paper about online EM (http://arxiv.org/abs/0712.4273). And I would hope that you could apply those improvements to the variational bayes extensions of EM (so that you could control clustering through the online optimization of the free energy). At least, that's what I wished I had time pursuing at the end of my PhD :) cheers, David ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2d-oct _______________________________________________ Scikit-learn-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
