A new approach to multiple class pattern classification with random matrices
Summary: We describe a new approach to multiple class pattern classification problems with noise and high dimensional feature spaces. The approach uses a random matrix \(X\) which has a specified distribution with mean \(M\) and covariance matrix \(r_{ij}(\Sigma_s+\Sigma_\varepsilon)\) between any two columns of \(X\). When \(\Sigma_\varepsilon\) is known, the maximum likelihood estimators of the expectation \(M\), correlation \(\Gamma\), and covariance \(\Sigma_s\) can be obtained. The patterns with high dimensional features and noise are then classified by a modified discriminant function according to the maximum likelihood estimation results. This new method is compared with a multilayer feed forward neural network approach on nine digit recognition tasks of increasing difficulty. Both methods achieved good results for those classification tasks, but the new approach was more effective and more efficient than the neural network method for difficult problems.
- scientific article; zbMATH DE number 3986452
- Decision theory classification of high-dimensional vectors based on small samples
- Correlation and High-Dimensional Consistency in Pattern Recognition
- scientific article; zbMATH DE number 3923896
- Point estimation for multi-spectral distributed random matrices
This page was built for publication: A new approach to multiple class pattern classification with random matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q951311)
