{"pk":31681,"title":"Representation of Temporal Patterns in Recurrent Networks","subtitle":null,"abstract":"In order to determine the manner in which\ntemporal patterns are represented in recurrent\nneural networks, networks trained on a vari-\nety of sequence recognition tasks are exam-\nined. Analysis of the state space of unit ac-\ntivations allows a direct view of the means em-\nployed by the network to solve a given prob-\nlem, and yields insight both into the class of\nsolutions these networks cfm produce and h o w\nthese will generalise to sequences outside the\ntraining set. This intuitive approach helps in\nassessing the potential of recurrent networks\nfor a variety of modelling problems.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[],"section":"Submitted Presentations","is_remote":true,"remote_url":"https://escholarship.org/uc/item/0qr0z21m","frozenauthors":[{"first_name":"Fred","middle_name":"","last_name":"Cummins","name_suffix":"","institution":"Indiana University","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"1993-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/31681/galley/22749/download/"}]}