Principal component analysis using neural network

The authors present their analysis of the differential equation dX ( t )/dt = AX ( t ) - X^T( t ) BX( t)X( t), where A is an unsymmetrical real matrix, B is a positive definite symmetric real matrix,X E Rn ; showing that the equation characterizes a class of continuous type full-feedback artificial neural network; We give the analytic expression of the solution; discuss its asymptotic behavior; and finally present the result showing that, in almost all cases, one and only one of following eases is true. 1. For any initial value X0∈R^n, the solution approximates asymptotically to zero vector. In thin cane, the real part of each eigenvalue of A is non-positive. 2. For any initial value X0 outside a proper subspace of R^n,the solution approximates asymptoticaUy to a nontrivial constant vector Y( X0 ). In this cane, the eigenvalue of A with maximal real part is the positive number λ=Ⅱ Y (X0)ⅡB^2 and Y (X0) is the corre-sponding eigenvector. 3. For any initial value X0 outsidea proper subspace of R^n, the solution approximates asymptotically to a non-constant periodic function Y( X0 , t ). Then the eigenvalues of A with maximal real part is a pair of conjugate complex numbers which can be computed.