Computational constrains often limit the practical applicability of coherent Bayes solutions to unsupervised sequential learning problems. These problems arise when attempts are made to learn about parameters on the basic of unclassified observations, each stemming from any one of k classes (k≥2). In this paper, the difficulties of the Bayes procedure will be discussed and existing approximate learning procedures will be reviewed for broad types of problems involving mixtures of probability densities. In particular a quasi-Bayes approximate learning procedure will be motivated and defined and its convergence properties will be reported for several special cases.
|Number of pages||13|
|Journal||Trabajos de Estadistica Y de Investigacion Operativa|
|State||Published - Feb 1980|
- Identifiable Mixtures
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty