Tea - Time - Talks

Time:          Fridays 3.30 PM

Location:   ICS 432


Schedule:

Jan. 09 '04.   Max Welling.

Markov Chain Sampling for Non-linear State Space Models Using Embedded Hidden Markov Models [NIPS-paper here: pdf] [techreport here: pdf] [powerpoint slides]

Radford Neal, Matt Beal, Sam Roweis, NIPS 2003.

Jan. 16 '04.   Eric Mjolsness.

Improving "Plate" Notation.

Jan. 23 '04.  Padhriac Smyth.

Modelling spatio-temporal rainfall data with graphical models.

Jan. 30 '04. Bozhean Bidyuk

The graph theoretical aspects of minimum weight w-cutset of a graph

Feb. 06 '04   Pierre Baldi.

On the Foundations of Machine Learning.

Feb. 13 '04

cancelled

Feb. 20 '04   David Van Dyke.

Rates of convergence of the EM algorithm and Gibbs samplers

Feb. 27 '04   Joshua O'Madadhain.

Predicting Connections in Networks

Mar. 05 '04

cancelled

Mar. 12 '04   Wayne Hayes, U. Maryland

Improving genome sequence assembly by using a better pair of glasses.

Mar. 19 '04   Albert Salah

Incremental Mixtures of Factor Analysers


Abstract:
A mixture of factor analyzer is a semiparametric density estimator that
performs
clustering and dimensionality reduction in each cluster (component)
simultaneously. It performs nonlinear dimensionality reduction by modeling the
density as a mixture of local linear models. The approach can be used for
classification by modeling each class-conditional density using a mixture model
and the complete data is then a mixture of mixtures. We propose an incremental
mixture of factor analysis algorithm where the number of components (local
models) in the mixture and the number of factors in each component (local
dimensionality) are determined adaptively. Our results on different pattern
classification tasks prove the utility of our approach and indicate that our
algorithms find a good trade-off between model complexity and accuracy.

Mar. 26 '04   

cancelled