AAAI Publications, Twenty-Third International FLAIRS Conference

Font Size: 
Performing Complex Associations Using a Feature-Extracting Bidirectional Associative Memory
Craig Leth-Steensen, Sylvain Chartier, Dominic Langlois, Marie-France Hebert

Last modified: 2010-05-06


Learning in Bidirectional Associative Memory (BAM) is typically based on Hebbian-type learning. Since Kosko’s paper on BAM in late 80s many improvements have been proposed. However, none of the proposed modifications allowed BAM to perform complex associative tasks that combine many-to-one with one-to-many associations. Even though BAMs are often deemed more plausible biologically, if they are not able to solve such mappings they will have difficulties establishing themselves as good models of cognition. This paper presents a BAM that can perform complex associations using only covariance matrices. It will be shown that this network can be trained to learn both the 2 and 3-bit parity problem and that the performance is increase by using a committee machine. The conditions that provide optimal learning performance within this latter network framework are explored along with some of its dynamical properties. The model is able to perform the nonlinear separation tasks while maintaining its properties.


Neural Networks; Bidirectional Associative Memory; nonlinear separable tasks; complex associations; committee machine

Full Text: PDF