Creating Perceptual Features Using a BAM-inspired Architecture

In this paper, it shown that the Feature-Extracting Bidirectional Associative Memory (FEBAM) can create its own set of perceptual features. Using a bidirectional associative memory (BAM)-inspired architecture, FEBAM 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. The model is tested by simulating prototype development in a noisy environment. Simulations show that the model fares particularly well compared to current neural PCA and independent component analysis (ICA) algorithms. Therefore, it is argued that the model possesses more cognitive explanative power than any other PCA/ICA algorithm or BAMs taken separately.