Department of Electrical Engineering, National Cheng Kung University
－Professor Jeen-Shing Wang－

Course Syllabi

Course Name Introduction to Neural Networks
Credits 3
Period Fall 2011
Objects This course provides students with insights into the fundamental concepts in the field of neural networks as an approach to the design of distributed intelligent and adaptive systems. In addition, this course will introduce students to the primary approaches to practical applications from a variety of fields such as pattern recognition, system identification, nonlinear prediction, and control as well.
Schedule
  1. Introduction
    • History
    • Artificial intelligence and neural networks  
    • Models of neurons
    • Network architectures
  2. Learning Methods
    • Supervised learning 
    • Unsupervised learning
  3. Multiple Layer Perceptrons
    • First order and second order optimization algorithms  
    • Structure/Parameter Learning 
    • Effects of /Heuristics for user predefined parameters
    • Examples/Homework
  4. Kohonen Self-Organizing Maps
    • Self-organizing map
    • The SOM algorithm
    • Learning vector quantization
  5. Kernel-based Networks
    • Radial-basis function networks
    • Interpolation/Regularization
    • Learning strategies
    • Support vector machines
    • Support vector clustering
  6. Dynamically Driven Recurrent Networks
    • Recurrent networks architectures
    • State-space Model
    • Learning algorithms
  7. Neuro-Fuzzy Systems
    • Preliminary of Fuzzy logic systems
    • Realization of fuzzy logic systems into connectionist networks
    • Existing neuro-fuzzy systems
  8. Existing Systematic/Unified Learning Frameworks
Texts & References
  1. Simon Haykin, Neural Networks: A Comprehensive Foundation, Prentice Hall, 1999.
  2. C.T. Lin and C.S.G. Lee, Neural Fuzzy Systems: A Neuro-Fuzzy Synergism to Intelligent Systems, Prentic-Hall PTR, 1996.
  3. IEEE Transactions on Neural Networks
  4. Proceedings of IEEE International Conference on Neural Networks
Lecture type  Lecture and class discussion
Grade 1. Term project (50%)
2. Paper critiques (25%)
3. Class participation (10%)
4. Paper presentation (15%)
Others  

　

Course Name Linear Algebra
Credits 3
Period Spring 2011
Objects This course will present the main concepts and terminology of linear algebra that play an essential role in mathematics and in many technical areas of modern society, such as computer science, engineering, physics, environmental science, economics, statistics, business management, and social sciences.
Schedule 1. Linear equations in linear algebra
2. Matrix algebra
3. Determinants
4. Vector spaces
5. Eigenvalues and eigenvectors
6. Orthogonality and least squares
7. Symmetric matrices and quadratic forms
Texts David C. Lay, Linear Algebra And Its Applications , 3rd Ed., 2002.
Lecture type Lecture and class discussion
Grade 1. Homework (10%)
2. Quizzes (10%)
3. Midterm exams (40%)
4. Final exam (40%)
Others  

　

Course Name  Applications of Optimization Theory
Credits  3
Period  Spring 2011
Objects This course provides students with a basic understanding of optimization problems including algorithms and search methods for optimization, iterative techniques (such as quasi-Newton, recursive least squares, genetic algorithm) and computational methods for unconstrained and constrained optimization. In addition, this course will introduce students to applications in systems and control problems, network training and parameter estimation.
Schedule
  1. Introduction
  2. Unconstrained optimization
  3. Least squares analysis
  4. Random search algorithms
  5. Linear programming
  6. Nonlinear constrained optimization
  7. Convex optimization
Texts E.K.P. Chong and S.H.Zak, An Introduction to Optimization, Second Edition, New York, NY: John Wiley & Sons, Inc. (Wiley-Interscience Series), 2001. (ISBN 0-471-39126-3)
Lecture type Lecture and class discussion in English
Grade 1. Homework due every 2 weeks: 15%
2. Two in-class exams: 50%
3. Final exam: 35%
Others