Adaptation from fixed weight dynamic networks

A characteristic often attributed to intelligent systems is adaptive behavior. For the purposes of this paper, we define adaptation as a system´s ability to recognize change through its sensed inputs and to appropriately adjust its behavior in response to the perceived change. This paper explores the notion that a time-lagged recurrent network architecture can be made to exhibit adaptive behavior after network training has been completed, i.e., to exhibit adaptation after its weights have been fixed and without any external mechanism to control its behavior