Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method

<table>Architecture of the recurrent neural
network. Solid arrows, dotted arrows, and boxes represent fixed connections,
adjustable connections, and network states, respectively.</table>

Advances in Artificial Neural Systems

fig1

Figure 1

Figure 1: Building Recurrent Neural Networks to Implement Multiple Attractor Dynamics Using the Gradient Descent Method 