This dissertation presents an introductory knowledge to computational neuroscience and major emphasize on the branch of computational neuroscience called Spiking Neural Networks (SNNs). SNNs are also called the third generation neural networks. It has become now a major field of Soft Computing. In this we talk about the temporal characteristics’ of neuron and studied the dynamics of it. We have presented SNNs architecture with fuzzy reasoning capability. Neuron selectivity is facilitated using receptive fields that enable individual neurons to be responsive to certain spike train frequencies and behave in a similar manner as fuzzy membership functions.
The network of SNNs consists of three layers that is input, hidden and output layer. The topology of this network is based on Radial basis Network, which can be regarded as universal approximators. The input layer receives the input in the form of frequency which produces the spikes through linear encoding. There is another method of encoding called Poisson encoding; this encoding is used where the data is large. The hidden layer use Receptive Field (RF) to process the input and thus it is frequency selective. The output layer is only responsible for learning. The learning is based on local learning.
The XOR classification problem is used to test the capabilities of the network. There is a problem of continuous updating of weight arises. This issue of weight is resolved by using STDP window and fuzzy reasoning.
The dissertation demonstrates how it is possible to obtain fuzzy reasoning capability from biological models of spiking neurons. The fuzzy spiking neural network implements fuzzy rules by configuration of receptive fields, antecedent conjunction with excitatory and inhibitory connections, and inferencing via a biologically plausible supervised learning algorithm. In this way, the resulting system utilizes a higher level of knowledge representation.
Inhaltsverzeichnis (Table of Contents)
- Abstract
- Introduction
- Fuzzy Spiking Neural Networks
- Introduction
- Biological Neural Networks
- Fuzzy Logic
- Spiking Neural Networks
- Fuzzy Spiking Neural Network (FSNN)
- FSNN Models
- Applications of FSNNs
- Control Systems
- Robotics
- Pattern Recognition
- Image Processing
- Signal Processing
- Medical Diagnosis
- Financial Applications
- Other Applications
- Challenges and Future Directions
- Conclusion
Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)
This dissertation aims to provide a comprehensive overview of Fuzzy Spiking Neural Networks (FSNNs), a novel type of neural network inspired by biological neural networks and incorporating elements of fuzzy logic and spiking neural networks. The dissertation explores the theoretical framework of FSNNs, their various models, and their potential applications in diverse fields.
- The integration of fuzzy logic and spiking neural networks in FSNNs
- The different models and architectures of FSNNs
- The potential applications of FSNNs in various domains, including control systems, robotics, and pattern recognition
- The challenges and future research directions in the field of FSNNs
- The potential benefits and advantages of FSNNs over traditional neural networks
Zusammenfassung der Kapitel (Chapter Summaries)
- Introduction: This chapter provides an overview of the dissertation, introducing the concept of Fuzzy Spiking Neural Networks and outlining the key objectives and research questions. It also discusses the motivation for exploring this novel type of neural network and its potential impact on various fields.
- Fuzzy Spiking Neural Networks: This chapter delves into the theoretical foundations of FSNNs, exploring the concepts of biological neural networks, fuzzy logic, and spiking neural networks. It explains how these concepts are integrated in FSNNs to create a new and powerful type of neural network.
- Applications of FSNNs: This chapter explores the potential applications of FSNNs in diverse fields, including control systems, robotics, pattern recognition, image processing, signal processing, medical diagnosis, financial applications, and other emerging areas. It discusses the advantages of FSNNs in addressing complex real-world problems and provides specific examples of their application.
- Challenges and Future Directions: This chapter examines the challenges and limitations of FSNNs, highlighting areas that require further research and development. It discusses potential solutions to these challenges and outlines promising future directions for the field of FSNNs.
Schlüsselwörter (Keywords)
This dissertation focuses on the intersection of fuzzy logic, spiking neural networks, and artificial intelligence. Key themes include the development of novel neural network models, the exploration of their capabilities and limitations, and the identification of promising applications in diverse fields. This research investigates the integration of fuzzy logic into spiking neural networks, aiming to enhance their robustness and accuracy in addressing complex problems.
Frequently Asked Questions
What are Spiking Neural Networks (SNNs)?
SNNs are considered the third generation of neural networks. They incorporate the concept of time into their operating model, using discrete spikes to transmit information, similar to biological neurons.
How is fuzzy logic integrated into Spiking Neural Networks?
Fuzzy reasoning is achieved by using receptive fields that act like fuzzy membership functions, allowing individual neurons to be selective to specific spike train frequencies.
What is the architecture of a Fuzzy Spiking Neural Network (FSNN)?
A typical FSNN consists of three layers: an input layer (receiving frequency data), a hidden layer (using receptive fields for selective processing), and an output layer responsible for learning.
What is the purpose of Poisson encoding in SNNs?
Poisson encoding is a method used to convert large datasets into spike trains, providing a biologically plausible way to represent input intensity through spike timing.
What are the potential applications for FSNNs?
FSNNs can be used in control systems, robotics, pattern recognition, medical diagnosis, image processing, and financial modeling due to their high-level knowledge representation.
- Quote paper
- Haider Raza (Author), 2011, Fuzzy Spiking Neural Networks, Munich, GRIN Verlag, https://www.grin.com/document/184731
