Course Synopsis
The complexity and dynamics of real-world problems, such as prediction, decision making for robots, control systems, large bioinformatics data processing, adaptive speech recognition and language acquisition, visual monitoring systems and multi-modal information processing, and intelligent agent based systems and adaptive agents on the web, require sophisticated methods and tools for building online, knowledge based intelligent systems. Such systems should be able to do the following: Learn adaptively, Dynamically create new modules, Memorize information, Interact continuously, Deal with knowledge.
Course Objectives
This 2 Day course will first discusses the fundamental principle of neural network and fuzzy logic, and then gives insight to tools available in MATLAB and SIMULINK to solve the complex and dynamic real-world problems. Numerous examples will be given to support the discussed theory.
Course Duration
2 Full Days
Who Must Attend
Engineers, researchers, scientists, postgraduate students, R&D staffs, and those who like to understand the principle of neural networks and fuzzy logic with their applications with MATLAB.
Prerequisite
Candidates must have experience with basic computer operation. Preferably attended our Technical Computing with MATLAB fundamental course.
Course Outline
Neural Network concepts
- Introduction
- Simple neuron model
- MATLAB representation of neural network
Type of Learning Methods
- Back propagation
- Least Square
- Steepest descent
Type of Neural Network
- Perceptrons
- Linear networks
- Multi layer perceptrons
- Self-organizing maps
Case Study
Fuzzy Logic Concepts
- Introduction
- Fuzzy Sets
- Membership functions
- Logical operations
- If-Then rules
Fuzzy inferences systems (FIS)
- Introduction
- Building FIS with Fuzzy GUI
- Working from the command line
- Application examples
Adaptive Neural-Fuzzy Inference Systems (ANFIS)
- Sugeno-type fuzzy inference
- The ANFIS editor GUI
- Working from the command line
- Application examples
Case Studies