Nonlinear Generalized Canonical Correlation Analysis by Neural Network Models
A method of K-set canonical correlation analysis capable of joint multivariate nonlinear transformations of data was proposed. The method consists of K nonlinear data transformation modules, each of which is a multi-layered feed-forward network, and one integrator module which combines information from the K transformation modules. The proposed method is useful for integrating information from K concurrent sources.
KeywordsHide Layer Response Category Output Activation Canonical Correlation Analysis Canonical Variate
Unable to display preview. Download preview PDF.
- Asoh, H., and Takechi, 0. (1994). An approximation of nonlinear canonical correlation analysis by multilayer perceptrons. ICANN 94 Proceedings of the International Conference on Artificial Neural Networks. New York: Springer, 713–716.Google Scholar
- Carroll, J.D. (1968). A generalization of canonical correlation analysis to three or more sets of variables. Proceedings of the 76th Annual Convention of the American Psychological Association, 227–228.Google Scholar
- Horst, P. (1961). Relations among m sets of measures. Psychometrika, 26, 129149.Google Scholar
- Marr, D. (1982). Vision: A computational investigation into the human representation and processing of visual information. San Francisco: Freeman.Google Scholar