creators_name: Phillips, S. type: journalp datestamp: 1998-12-10 lastmod: 2011-03-11 08:54:01 metadata_visibility: show title: Are feedforward and recurrent networks systematic? Analysis and implications for a connectionist cognitive architecture ispublished: pub subjects: cog-psy subjects: comp-sci-neural-nets subjects: phil-mind full_text_status: public keywords: strong systematicity, connectionism, classicism, feedforward network, recurrent network, tensor network, weight sharing, local representation, compositionality, association, relation abstract: Human cognition is said to be systematic: cognitive ability generalizes to structurally related behaviours. The connectionist approach to cognitive theorizing has been strongly criticized for its failure to explain systematicity. Demonstrations of generalization notwithstanding, I show that two widely used networks (feedforward and recurrent) do not support systematicity under the condition of local input/output representations. For a connectionist explanation of systematicity, these results leave two choices, either: (1) develop models capable of systematicity under local input/output representations; or (2) justify the choice of similarity-based (nonlocal) component representations sufficient for systematicity. date: 1998 date_type: published publication: Connection Science volume: 10 number: 2 pagerange: 137-160 refereed: TRUE citation: Phillips, S. (1998) Are feedforward and recurrent networks systematic? Analysis and implications for a connectionist cognitive architecture. [Journal (Paginated)] document_url: http://cogprints.org/520/2/systemffn-rn.ps