Fuzzy Art Map

Only available on StudyMode
  • Topic: Artificial neural network, Machine learning, Fuzzy logic
  • Pages : 4 (1054 words )
  • Download(s) : 68
  • Published : December 19, 2012
Open Document
Text Preview
FUZZY ARTMAP & ITS APPLICATION
Priyatam Kumar Kanak
Reg no-10906969, Roll no- REE026A27
Department of Electronics Engineering
Lovely Professional University

Abstract- Fuzzy ARTMAP is an architecture which is based on the synthesis of fuzzy logic and adaptive resonance theory (ART) neural networks, by exploiting a close formal similarity between the computations of fuzzy subsets and ART category, resonance and learning. It is a powerful neural network in application of prediction and classifier.

Keywords- ART,ARTMAP, Fuzzy Logic, FAM (Fuzzy ARTMAP), SBFS (Sequential Backward Floating Search)

I.
Introduction- The basic ART architecture was developed by Grossberg and Carpenter. The major innovation of ART is the use of “expectations.” In ART architecture, each input pattern is presented to the network which is compared with the prototype vector that is most closely matches (expectation). If the match between the prototype and the input vector is not adequate a new prototype is selected. In this way previously learned memories are not eroded by new learning. It consists of three parts: Layer2 (L2) to Layer1 (L1) expectations, the orienting system and gain control.

In the ART networks, learning also occurs in a set of feedback connections from layer2 to layer1. These connections are outstars which perform pattern recall. When a node in layer2 is activated, this reproduces a prototype pattern (the expectation) at layer1. The layer1 then performs a comparison between the expectation and the input pattern. When the expectation and the input pattern are not closely matched, the orienting subsystem causes a reset in layer2. This reset disables the current winning neuron, and the current expectation is removed. A new expectation is then performed in layer2, while the pervious winning neuron is disabled. The new winning neuron in layer2 projects a new expectation to layer1 through the L2-L1 connections. This process continues until the L2-L1...
tracking img