From ANN to Biomimetic Information Processing
Artificial neural networks (ANN) are useful components in today’s data analysis toolbox. They were initially inspired by the brain but are today accepted to be quite different from it. ANN typically lack scalability and mostly rely on supervised learning, both of which are biologically implausible features. Here we describe and evaluate a novel cortex-inspired hybrid algorithm. It is found to perform on par with a Support Vector Machine (SVM) in classification of activation patterns from the rat olfactory bulb. On-line unsupervised learning is shown to provide significant tolerance to sensor drift, an important property of algorithms used to analyze chemo-sensor data. Scalability of the approach is illustrated on the MNIST dataset of handwritten digits.
KeywordsSupport Vector Machine Artificial Neural Network Olfactory Bulb Hybrid Algorithm Handwritten Digit
Unable to display preview. Download preview PDF.
- Johansson, C.: An attractor memory model of neocortex. School of Computer Science and Communication. Stockholm, Royal Institute of Technology, Sweden. PhD (2006)Google Scholar
- Johansson, C., Lansner, A.: A hierarchical brain-inspired computing system. In: International Symposium on Nonlinear Theory and its applications (NOLTA), Bologna, Italy, pp. 599–603 (2006b)Google Scholar
- Leon, M., Johnson, B.A.: Glomerular activity response archive (2006), http://leonserver.bio.uci.edu
- Ma, J., Zhao, Y., Ahalt, S., Eads, D.: OSU-SVM for Matlab (2006), http://svm.sourceforge.net
- Marco, S., Lansner, A., Gutierrez Galvez, A.: Exploratory analysis of the rat olfactory bulb activity. Abstract. ECRO 2006, Granada, Spain (2006)Google Scholar
- Quinlan, P.: Connectionism and psychology. A psychological perspective on connectionist research. New York, Harvester, Whaetsheaf (1991)Google Scholar
- Steinert, R., Rehn, M., Lansner, A.: Recognition of handwritten digits using sparse codes generated by local feature extraction methods. In: 14th European Symposium on Artificial Neural Networks (ESANN) 2006, Brugge, Belgium, pp. 161–166 (2006)Google Scholar
- Thorpe, S., Imbert, M.: Biological constraints on connectionist modelling. In: Pfeiffer, R., Berlin, E. (eds.) Connectionism in Perspective. Springer, Berlin (1989)Google Scholar
- Young, F.W.: Multidimensional scaling. Encyclopedia of Statistical Sciences 5, 649–659 (1985)Google Scholar