{"pk":29503,"title":"Relation learning in a neurocomputational architecture supports cross-domaintransfer","subtitle":null,"abstract":"Humans readily generalize, applying prior knowledge to novelsituations and stimuli. Advances in machine learning have be-gun to approximate and even surpass human performance, butthese systems struggle to generalize what they have learnedto untrained situations. We present a model based on well-established neurocomputational principles that demonstrateshuman-level generalisation. This model is trained to play onevideo game (Breakout) and performs one-shot generalisationto a new game (Pong) with different characteristics. The modelgeneralizes because it learns structured representations that arefunctionally symbolic (viz., a role-filler binding calculus) fromunstructured training data. It does so without feedback, andwithout requiring that structured representations are specifieda priori. Specifically, the model uses neural co-activation todiscover which characteristics of the input are invariant and tolearn relational predicates, and oscillatory regularities in net-work firing to bind predicates to arguments. To our knowledge,this is the first demonstration of human-like generalisation ina machine system that does not assume structured representa-tions to begin with.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[{"word":"predicate learning; generalisation; neural net-works; symbolic-connectionism; neural oscillations"}],"section":"Forms of Learning","is_remote":true,"remote_url":"https://escholarship.org/uc/item/35v29557","frozenauthors":[{"first_name":"Leonidas","middle_name":"A. A.","last_name":"Doumas","name_suffix":"","institution":"University of Edinburgh","department":""},{"first_name":"Guillermo","middle_name":"","last_name":"Puebla","name_suffix":"","institution":"University of Edinburgh","department":""},{"first_name":"Andrea","middle_name":"E.","last_name":"Martin","name_suffix":"","institution":"Max Planck Institute for Psycholinguistics","department":""},{"first_name":"John","middle_name":"E.","last_name":"Hummel","name_suffix":"","institution":"University of Illinois","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"2020-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/29503/galley/19363/download/"}]}