We realize a multicategory Bayesian classifier by a three-layer neural network having rather a small number of hidden layer units. The state-conditional probability distributions are supposed to be multivariate normal distributions. The network has direct connections between the input and output layers. Its outputs are monotone mappings of posterior probabilities. Hence, they can be used as discriminant functions and, in addition, the posterior probabilities can be easily retrieved from the outputs. © Springer-Verlag Berlin Heidelberg 2003.
CITATION STYLE
Ito, Y., & Srinivasan, C. (2003). Multicategory Bayesian decision using a three-layer neural network. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2714, 253–261. https://doi.org/10.1007/3-540-44989-2_31
Mendeley helps you to discover research relevant for your work.