Introduction

This tutorial is an accompanying document to the computer code for approximate information discriminant analysis. The details of the method can be found in [1] and the computer code is written in MATLAB^TM. The code consists of several basic functions:

1. aida.m
2. sqrtmat.m
3. invsqrtmat.m
4. logmat.m

aida.m returns the optimal feature extraction matrix, T^*. The input arguments of aida.m allow for choice of size of feature space. Type help aida in MATLAB^TMcommand prompt to learn more about this function.

sqrtmat.m is a function that extracts squareroot of any covariance matrix A, as defined in [1]. This function is an auxiliary function in aida.m.

invsqrtmat.m is a function that extracts the squareroot of inverse of any covariance matrix, A as defined in [1]. This function is an auxiliary function in aida.m.

logmat.m is a function that extracts the logarithm of any covariance matrix A, as defined in [1]. This function is an auxiliary function in aida.m.