Mar 29, 2024  
2014-2015 Graduate Academic Catalog 
    
2014-2015 Graduate Academic Catalog [ARCHIVED CATALOG]

Add to Portfolio (opens a new window)

EE 584 - Neural Networks

3 lecture hours 0 lab hours 3 credits
Course Description
This course introduces students to the basic concepts of modeling and simulating adaptive and learning systems using neural networks. The underlying concepts of neural networks are introduced, as well as a number of common topologies and learning rules used in neural networks. Students gain firsthand experience of neural networks through computer assignments and a short research project. (prereq: CS-2510 or equivalent, MA-343 or MA-383)
Course Learning Outcomes
Upon successful completion of this course, the student will be able to:
• Describe the origins of neural networks.
• Describe how neural networks store information.
• Describe the basic configurations of neural networks.
• Describe and implement various learning algorithms for neural networks.
• Develop and simulate simple neural networks.
• Discuss engineering problems to which neural networks may be a suitable solution.
• Use commercially available neural network development tools.
• Perform basic literature searches and prepare short presentations.
Prerequisites by Topic
• High-level language programming.
• Basic numerical methods.
• Advanced calculus (Taylor Series, gradients, etc.).
Course Topics
• Introduction to neural networks, problems, terminology, MATLAB toolbox (4 classes)
• Data gathering and formatting (2 classes)
• Linear perceptron, learning rules, training styles, convergence of weights (3 classes)
• Multilayer networks, construction and transfer functions (3 classes)
• Backpropagation, training algorithms and associated mathematics (4 classes)
• Choosing relevant inputs and pruning (1 class)
• Special topics (7 classes)
• Project workshops (5 classes)
• Student presentations (3 classes)
Coordinator
Sheila Ross



Add to Portfolio (opens a new window)