Title :
Noisy recurrent neural networks: the discrete-time case
Author :
Olurotimi, Oluseyi ; Das, Soumitra
Author_Institution :
Dept. of Electr. & Comput. Eng., George Mason Univ., Fairfax, VA, USA
fDate :
9/1/1998 12:00:00 AM
Abstract :
Fingings concerning the behavior of continuous-time recurrent neural networks (RNN) are specialized to the discrete-time (so-called time-lagged recurrent) case. Uniform boundedness of the first and second moment sequences of the trajectory is established. For practical design purposes, estimates and bounds on the bias and variance sequences of the stochastic discrete-time RNN are derived. A script file for estimating the variance bounds for specified designs is provided
Keywords :
complete computer programs; noise; recurrent neural nets; MATLAB script file; bias sequence; discrete-time noisy recurrent neural networks; stochastic discrete-time RNN; time-lagged recurrent neural nets; trajectory moment sequences; uniform boundedness; variance sequence; Additive white noise; Computer aided software engineering; Control systems; Difference equations; Differential equations; Neurons; Nonlinear control systems; Recurrent neural networks; Stochastic resonance; White noise;
Journal_Title :
Neural Networks, IEEE Transactions on
