PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Learning mixture models - courseware for finite mixture distributions of multivariate Bernoulli distributions
Jaakko Hollmen and Tapani Raiko
In: In Proceedings of Teaching Machine Learning - workshop on open problems and new directions, 6 May - 7 May 2008, Saint-Étienne, France.

Abstract

Teaching of machine learning should aim at the readiness to understand and implement modern machine learning algorithms. Towards this goal, we often have course exercises involving the student to solve a practical machine learning problem involving a reallife data set. The students implement the programs of machine learning methods themselves and gain deep insight on the implementation details of the method. The downside of this approach is that time is devoted on implementation aspects rather than machine learning. Complementary to this approach, we have designed a machine learning course exercise on a ready implementation of the Expectation-Maximization (EM) algorithm for finite mixture distributions of multivariate Bernoulli distributions. We describe BernoulliMix — a program package with a set of teaching examples and exercises and report on the preliminary experiences in our class of machine learning students. The BernoulliMix package will be available under a liberal open source license.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Additional Information:Software available from the BernoulliMix homepage at: http://www.cis.hut.fi/jhollmen/BernoulliMix/
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:4203
Deposited By:Jaakko Hollmen
Deposited On:21 November 2008