Thank you all! However, what Sturla wrote is now out of my understanding. One more question. It seems also to me that Naive Bayes classifiers also do not need data to be scaled. Am I correct?
Best Regards, Yury Zhauniarovich On 4 June 2015 at 20:55, Sturla Molden <sturla.mol...@gmail.com> wrote: > On 04/06/15 20:38, Sturla Molden wrote: > > > Component-wise EM (aka CEM2) is a better way of avoiding the singularity > > disease, though. > > The traditional EM for a GMM proceeds like this: > > while True: > > global_estep(clusters) > > for c in clusters: > mstep(c) > > This is inherently unstable. Several clusters can become > near-singular in the M-step before there is an E-step > to redistribute the weights. You can get a "cascade of > singularities" where the whole GMM basically dies. Even > if you bias the diagonal of the covariance you still > have the basic algorithmic problem. > > CEM2 proceeds like this: > > while True: > for c in clusters: > estep(c) > mstep(c) > > This improves stability enormously. When a cluster becomes > singular, the memberships are immediately redistributed. > Therefore you will not get a "cascade of singularities" > where the whole GMM basically dies. > > > Sturla > > > > ------------------------------------------------------------------------------ > _______________________________________________ > Scikit-learn-general mailing list > Scikit-learn-general@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/scikit-learn-general >
------------------------------------------------------------------------------
_______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
