Convergence Analysis of Continuous-Time Neural Networks
The energy function of continuous-time neural network has been analyzed for testing the existence of stationary points and the global convergence of network. The energy function always has only one stationary point which is a saddle point in the unconstrained space when the total conductance of neuron’s input is zero (G i = 0). However, the stationary points exist only inside the hypercube Rn ∈[0,1] when the total conductance of neuron’s input is not zero (G i ≠ 0). The Hessian matrix of the energy function is used for testing the global convergence of the network.
Unable to display preview. Download preview PDF.
- 3.Friedberg, S., Insel, A.: Introduction to Linear Algebra with Application, 1st edn. Prentice-Hall, Englewood Cliffs (1986)Google Scholar
- 5.Zurada, J.M.: Introduction to Artificial Neural Systems, 1st edn. West, St. Paul MN (1992)Google Scholar
- 7.Golub, G.H., Van Loan, C.F.: Matrix Computations, 3rd edn. John Hopkins Univ. Press, Baltimore (1996)Google Scholar