Research indexJournal article
Toward automatic constructive learning
Shultz
Behavioral and Brain Sciences1 Jun 2008
NobleIDark:/48914/w1/20260525/CCD2B417DOI10.1017/s0140525x08004196
Toward automatic constructive learning
Shultz · Behavioral and Brain Sciences · 2008 · Subscription
NobleIDark:/48914/w1/20260525/CCD2B417DOI10.1017/s0140525x08004196
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
Abstract Neuroconstructivist modeling can be usefully extended with algorithms that build their own topology and recruit existing knowledge, effectively constructing a hierarchy of network modules. Possible benefits include allowing abilities to emerge naturally, in a way that affords objective study, deeper insights, and more rapid progress, and provides more serious consideration of the implications of constructivism.