Boolean Factor Analysis by Expectation-Maximization Method

0368469 - ÚI 2013 RIV DE eng C - Conference Paper (international conference)
Frolov, A. A. - Húsek, Dušan - Polyakov, P.Y.
Boolean Factor Analysis by Expectation-Maximization Method.
Proceedings of the Third International Conference on Intelligent Human Computer Interaction IHCI 2011. Heidelberg: Springer, 2013 - (Kudělka, M.; Pokorný, J.; Snášel, V.; Abraham, A.), s. 243-254. Advances in Intelligent Systems and Computing, 179. ISBN 978-3-642-31602-9. ISSN 2194-5357.
[IHCI 2011. International Conference on Intelligent Human Computer Interaction /3./. Prague (CZ), 29.08.2011-31.08.2011]
R&D Projects: GA ČR GAP202/10/0262; GA ČR GA205/09/1079
Grant - others:GA MŠk(CZ) ED1.1.00/02.0070
Program: ED
Institutional research plan: CEZ:AV0Z10300504
Keywords : neural networks * hidden pattern search * Boolean factor analysis * generative model * information redundancy * exceptation-maximization
Subject RIV: IN - Informatics, Computer Science

Boolean factor analysis is one of the most efficient methods to reveal and to overcome informational redundancy of high-dimensional binary signals. In the present study, we introduce new Expectation-Maximization method which maximizes the likelihood of Boolean factor analysis solution. Using the so-called bars problem benchmark, we compare efficiencies of the proposed method with Dendritic Inhibition neural network.
Permanent Link: http://hdl.handle.net/11104/0202803