C++ Neural Networks and Fuzzy Logic: Preface

The Network for ART1

The neural network for the adaptive resonance theory or ART1 model consists of the following:

•  A layer of neurons, called the F

1

 layer (input layer or comparison layer)

•  A node for each layer as a gain control unit

•  A layer of neurons, called the F

2

 layer (output layer or recognition layer)

•  A node as a reset unit

•  Bottom−up connections from F

1

 layer to F

2

 layer

•  Top−down connections from F

2

 layer to F

1

 layer

•  Inhibitory connection (negative weight) form F

2

 layer to gain control

•  Excitatory connection (positive weight) from gain control to a layer

C++ Neural Networks and Fuzzy Logic:Preface

Chapter 10 Adaptive Resonance Theory (ART)

199

---

•  Inhibitory connection from F

1

 layer to reset node

•  Excitatory connection from reset node to F

2

 layer