Towards a constructivist methodology: learning constructions byintegrating in situ representations and productivity

摘要

The ability to learn constructions may be important for the development of a self-organizingarchitecture for artificial general intelligence. Constructions are structural relations between morespecific or more abstract conceptual representations. They can be derived from the processesof alignment, collocations and distributed equivalences. An architecture that integrates in situgrounded representations with cognitive productivity is ideally suited to learn constructions.This paper described such an architecture, based on neuronal assembly structures and neuronal’blackboards’ for grounded compositional representations. The paper outlines how constructionscould be learned in such an architecture and how the architecture could eventually develop into anautonomous self-organizing architecture for artificial general intelligence.