Every indexed document

because the main interest of their review is in monolithic systems. We

but playing a clearly derived and secondary role. We prefer

language and usage. “Language” refers to the type of language used in the symbolic

a paper that gives an elegant categorization of neural-symbolic AI systems. Figure

or using logical or neural net components

and globalist versus localist knowledge representation.

real-time interaction between the internal

encapsulated in black boxes and communicating results with each other.”

nearly all the existing hybrid architectures have severe shortcomings which

intended to mimic the qualitative structure of

4D/RCS is not foundationally learning-based but relies

and the same three cross-connected hierarchies as DeSTIN

e.g. it has the same hierarchical pattern recognition

after a significant time of development it has not proved significantly

much like Hofstadter’s earlier Copycat architecture [Hof95]

but the overall cognitive architecture is simplistic compared

learning and reasoning are carried out by algorithms that seem unlikely to

handling more advanced aspects like language and reasoning

but it deals most thoroughly with “lower

self-organizing to collectively carry out tasks such as perception