ECE 62900 - Introduction to Neural Networks

Course Details

Credits: 3

Areas of Specialization:

  • Communications, Networking, Signal & Image Processing

Counts as:

Normally Offered:

Fall - even years

Campus/Online:

On-campus and online

Catalog Description:

Information processing with neural networks, biological and engineering implications, learning algorithms, current neural network models and architectures, implementation topics, applications in areas such as signal/image processing, pattern recognition, optimization, simulation, system identification, nonlinear prediction, communications and control.

Required Text(s):

  1. Neural Networks and Learning Machines , 3rd Edition , S. Haykin , Prentice-Hall , 2008 , ISBN No. 0131471392

Recommended Text(s):

None.

Lecture Outline:

Weeks Topic
2.0 1. Introduction to Artificial Neural Networks A. Overview and Fundamental Concepts B. Historical Development C. Basic Models and Learning Rules of ANNS D. Distributed Representations E. Linear Threshold Elements and the Perceptron
1.0 2. Feedforward and Multistage Networks A. One Stage and Multistage Feedforward Networks B. Introduction to Sigmoidal, Radial Basis, and Other Activation Functions
3.0 3. Supervised Learning of Feedforward Networks A. Discriminant Functions B. The Perceptron Learning Algorithm C. The Least Mean Square (LMS) Algorithm and the Adaline D. The Backpropogation Learning Algorithm E. Convergence Analysis F. Optimal Choices of Learning Parameters G. Generalization Properties
3.0 4. Recurrent Networks A. Hopfield Networks B. Associative Memory C. Relations to Liapunov Functions and Stability of Nonlinear Systems D. Optimization Problems E. Recurrent Backpropogation Networks F. Avalanche Networks G. Reinforcement Learning
3.0 5. Unsupervised Learning Networks A. Competitive Learning and Other Unsupervised Learning Rules B. Self-organizing Feature Maps C. Adaptive Resonance Theory and the Leader Algorithm D. Principal Component Analysis E. Other Unsupervised Learning Networks
2.0 6. Applications A. Signal/Image Processing and Recognition B. System Identification C. Nonlinear Prediction D. Control Applications E. Intelligence Applications F. Communications Applications G. Fault Diagnosis H. Optimization
1.0 7. Exams

Assessment Method:

none