FEBAM: A Feature-Extracting Bidirectional Associative Memory

In this paper, a new model that can ultimately create its own set of perceptual features is proposed. Using a bidirectional associative memory (BAM)-inspired architecture, the resulting model inherits properties such as attractor-like behavior and successful processing of noisy inputs, while being able to achieve principal component analysis (PCA) tasks such as feature extraction and dimensionality reduction. The model is tested by simulating image reconstruction and blind source separation tasks. Simulations show that the model fares particularly well compared to current neural PCA and independent component analysis (ICA) algorithms. It is argued the model possesses more cognitive explanative power than any other nonlinear/linear PCA and ICA algorithm.

[1]  J J Hopfield,et al.  Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[2]  Sun-Yuan Kung,et al.  Principal Component Neural Networks: Theory and Applications , 1996 .

[3]  Erkki Oja,et al.  The nonlinear PCA criterion in blind source separation: Relations with other approaches , 1998, Neurocomputing.

[4]  Songcan Chen,et al.  A novel multi-valued BAM model with improved error-correcting capability , 2003 .

[5]  Sébastien Hélie,et al.  Are unsupervised neural networks ignorant? Sizing the effect of environmental distributions on unsupervised learning , 2006, Cognitive Systems Research.

[6]  Bart Kosko,et al.  Unsupervised learning in noise , 1990, International 1989 Joint Conference on Neural Networks.

[7]  Dominique Valentin,et al.  Eigenfeatures as intermediate-level representations: The case for PCA models , 1998, Behavioral and Brain Sciences.

[8]  Richard S. Sutton,et al.  Learning to predict by the methods of temporal differences , 1988, Machine Learning.

[9]  BART KOSKO,et al.  Bidirectional associative memories , 1988, IEEE Trans. Syst. Man Cybern..

[10]  Robert L. Goldstone,et al.  The development of features in object concepts , 1998, Behavioral and Brain Sciences.

[11]  T. Ens,et al.  Blind signal separation : statistical principles , 1998 .

[12]  Erkki Oja,et al.  Neural Networks, Principal Components, and Subspaces , 1989, Int. J. Neural Syst..

[13]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[14]  Robert Proulx,et al.  NDRAM: nonlinear dynamic recurrent associative memory for learning bipolar and nonbipolar correlated patterns , 2005, IEEE Transactions on Neural Networks.

[15]  G. Murphy,et al.  The Big Book of Concepts , 2002 .

[16]  Mohamad H. Hassoun.  Dynamic heteroassociative neural memories , 1989, Neural Networks.

[17]  Mounir Boukadoum,et al.  A bidirectional heteroassociative memory for binary and grey-level patterns , 2006, IEEE Transactions on Neural Networks.

[18]  Giovanni Costantini,et al.  Neural associative memory storing gray-coded gray-scale images , 2003, IEEE Trans. Neural Networks.

[19]  Jacek M. Zurada,et al.  Generalized Hopfield networks for associative memories with multi-valued stable states , 1996, Neurocomputing.

[20]  Mounir Boukadoum,et al.  A sequential dynamic heteroassociative memory for multistep pattern recognition and one-to-many association , 2006, IEEE Transactions on Neural Networks.

[21]  Chua-Chin Wang,et al.  Capacity analysis of the asymptotically stable multi-valued exponential bidirectional associative memory , 1996, IEEE Trans. Syst. Man Cybern. Part B.