Friday, December 19, 2014
Further use of simple parallel processing with Asa H
The Asa H architecture consists of a hierarchical memory assembled out of clustering modules and feature detectors. I have done some feature extraction outside of the main Asa H program using an autoassociative neural network trained on the various cases from the Asa H casebase. The hidden layer of the autoassociator network provides the feature detectors. These networks require considerable time to train by backpropagation. A half dozen or so networks can be trained in parallel on individual computers.