Dimension reduction for hidden Markov models using the suficiency approach

TOMASSI, DIEGO; FORZANI, LILIANA; MILONE, DIEGO; COOK, R. DENNIS

Córdoba

Simposio; Simposio Argentino de Inteligencia Artificial - JAIIO; 2011

SADIO- UTN Fac. Regional Córdoba

Dimension reduction is often included in pattern recognizers based on hidden Markov models to lower the size of the models to estimate. Commonly used methods are heuristic in nature and do not take care of information retention after projection. In this paper we present a new method based on the approach of sufficient dimension reductions. It explicitly accounts for all the discriminative information available in the original features, while using a minimum number of linear combinations of them. We review the underlying theory and present an algorithm for practical implementation of the proposed method. In the experimental side, we use simulations to illustrate its advantages over widely-used existing alternatives. In particular, we show that it performs as good as existing techniques when data is optimal according to the assumptions of those techniques, but significantly better for heteroscedastic data with no special structure on the covariance matrix.