Biased ART: A Neural Architecture that Shifts Attention Toward Previously Disregarded Features Following an Incorrect Prediction


Autoria(s): Carpenter, Gail A.; Gaddam, Sai Chaitanya
Data(s)

14/11/2011

14/11/2011

01/04/2009

Resumo

Memories in Adaptive Resonance Theory (ART) networks are based on matched patterns that focus attention on those portions of bottom-up inputs that match active top-down expectations. While this learning strategy has proved successful for both brain models and applications, computational examples show that attention to early critical features may later distort memory representations during online fast learning. For supervised learning, biased ARTMAP (bARTMAP) solves the problem of over-emphasis on early critical features by directing attention away from previously attended features after the system makes a predictive error. Small-scale, hand-computed analog and binary examples illustrate key model dynamics. Twodimensional simulation examples demonstrate the evolution of bARTMAP memories as they are learned online. Benchmark simulations show that featural biasing also improves performance on large-scale examples. One example, which predicts movie genres and is based, in part, on the Netflix Prize database, was developed for this project. Both first principles and consistent performance improvements on all simulation studies suggest that featural biasing should be incorporated by default in all ARTMAP systems. Benchmark datasets and bARTMAP code are available from the CNS Technology Lab Website: http://techlab.bu.edu/bART/.

SyNAPSE program of the Defense Advanced Projects Research Agency (Hewlett-Packard Company, subcontract under DARPA prime contract HR0011-09-3-0001; HRL Laboratories LLC, subcontract #801881-BS under DARPA prime contract HR0011-09-C-0001); Science of Learning Centers program of the National Science Foundation (SBE-0354378)

Identificador

http://hdl.handle.net/2144/1967

Idioma(s)

en_US

Publicador

Boston University Center for Adaptive Systems and Department of Cognitive and Neural Systems

Relação

BU CAS/CNS Technical Reports;CAS/CNS-TR-2009-003

Direitos

Copyright 2009 Boston University. Permission to copy without fee all or part of this material is granted provided that: 1. The copies are not made or distributed for direct commercial advantage; 2. the report title, author, document number, and release date appear, and notice is given that copying is by permission of BOSTON UNIVERSITY TRUSTEES. To copy otherwise, or to republish, requires a fee and / or special permission.

Boston University Trustees

Palavras-Chave #Adaptive Resonance Theory (ART) #ART #ARTMAP #Featural biasing #Supervised learning #Top-down/bottom-up interactions
Tipo

Technical Report