This site has been permanently archived. This is a static copy provided by the University of Southampton.

@misc{cogprints520, volume = {10}, number = {2}, title = {Are feedforward and recurrent networks systematic? Analysis and implications for a connectionist cognitive architecture}, author = {S. Phillips}, year = {1998}, pages = {137--160}, journal = {Connection Science}, keywords = {strong systematicity, connectionism, classicism, feedforward network, recurrent network, tensor network, weight sharing, local representation, compositionality, association, relation}, url = {http://cogprints.org/520/}, 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.} }