The Divergent Autoencoder (DIVA) Account of Human Category Learning
Kenneth J. Kurtz · eScholarship (California Digital Library) · 2005
The DIVA network model is introduced based on the novel computational principle of divergent autoencoding. DIVA produces excellent fits to classic data sets from Shepard, Hovland & Jenkins (1961) and Medin & Schafffer (1978). DIVA is also resistant to catastrophic interference. Such results have not previously been demonstrated by a model that is not committed to both localist coding of exemplars (or exceptions) and the use of an explicit selective attention mechanism.