Nonlinear system on-line identification via dynamic neural networks is stud
ied in this paper. The main contribution of the paper is that the passivity
approach is applied to access several new stable properties of neuro ident
ification. The conditions for passivity, stability, asymptotic stability, a
nd input-to-state stability are established in certain senses. We conclude
that the gradient descent algorithm for weight adjustment is stable in an L
-infinity sense and robust to any bounded uncertainties.