Schneider S., Abdel-Fattah A., Angerer B. & Weber F. (2013) Model construction in general intelligence. In: Kühnberger K.-U., Rudolph S. & Wang P. (eds.) Proceedings of the AGI 2013. Springer, Berlin: 109–118. Fulltext at https://cepa.info/937
Model construction in general intelligence.
In: Kühnberger K.-U., Rudolph S. & Wang P. (eds.) Proceedings of the AGI 2013. Springer, Berlin: 109–118.
Fulltext at https://cepa.info/937
In this conceptual paper we propose a shift of perspective for parts of AI – from search to model construction. We claim that humans construct a model of a problem situation consisting of only a few, hierarchically structured elements. A model allows selective exploration of possible continuations and solutions, and for the flexible instantiation and adaptation of concepts. We underpin our claims with results from two protocol studies on problem-solving imagery and on the inductive learning of an algorithmic structure. We suggest that a fresh look into the small-scale construction processes humans execute would further ideas in categorization, analogy, concept formation, conceptual blending, and related fields of AI. Relevance: In accordance with von Glasersfeld’s postulate that knowledge is actively built up by the cognizing subject, the paper emphasizes the constructive nature of human intelligence. While problem space models in AI also partly reflect this, the metaphorical language of problem space search lends itself to epistemological misinterpretations.