FuzzySystems与NeuralNetwork的整合.doc

PAGE 2 PAGE 9 Fuzzy Systems與Neural Network的整合 I. Introduction ☆ Fuzzy Logic Techniques Most often use verbal and Linguistic Information from experts ☆ Neural Networks Extract Information from Systems to be Learned or Controlled To obtain the benefits of both fuzzy Systems Neural Networks Fuzzy Systems:-Higher Level Humanlike IF-THEN rules thinking reasoning Ease of Incorporating Expert Knowledge Neural Networks:-Lower Level Learning Meilihei Optimization Meilihei Connectionist Structures 1. Neural Fuzzy Systems 為主 2. Fuzzy Neural Network 為主 3. Fuzzy-Neural Hybrid Systems 綜合體 1. Neural Fuzzy Systems: The use of Neural Networks as fools in Fuzzy Models 2. Fuzzy Neural Networks: Fuzzification of Conventional Neural Network Models 3. Fuzzy-Neural Hybrid Systems: Incorporating Fuzzy Technologies and Neural Networks Into Hybrid Systems. 1. Neural Fuzzy Systems Providing Fuzzy Systems with the kind of automatic tuning methods typical of Neural Networks. But Without Altering Their Functionality (e.g. Fuzzification, Defuzzification, Inference Engine Fuzzy Logic Base) Augmenting Numerical Processing of Fuzzy Sets Since Neural Fuzzy Systems are Inherently Fuzzy Logic Systems. They are mostly used in Control Applications 2. Fuzzy Neural Networks: Retain the basic properties and architecture of neural networks and simply fuzzify some of their Elements e.g. a Crisp Neuron can become fuzzy and the response of the neuron to the lower- layer activation signal can be of a fuzzy relation type rather than a signed type Domain knowledge becomes formalized In forms of fuzzy sets and afterward can be applied to enhance the learning algorithms of the neural networks or augment their interpretation capabilities Since fuzzy neural networks are inherently neural networks, they are mostly used in pattern recognition applications 3. Fuzzy-Neural Hybrid Systems: Both play a key role in hybrid Systems ANFIS: Adaptive Network-Based Fuzzy Inferenc