作者: Paul S. Rosenbloom , Abram Demski , Volkan Ustun
DOI: 10.1007/978-3-319-41649-6_9
关键词: Graphical model 、 Feedforward neural network 、 Sigma 、 Artificial intelligence 、 Artificial neural network 、 Computer science 、 Cognitive architecture 、 Architecture 、 Semantics (computer science) 、 Extension (predicate logic) 、 Theoretical computer science
摘要: The status of Sigma’s grounding in graphical models is challenged by the ways which their semantics has been violated while incorporating rule-based reasoning into them. This led to a rethinking what goes on its architecture, with results that include straightforward extension feedforward neural networks (although not yet learning).