Artificial Neural Networks

Course Nos. 0909-560-01, 0909-454-01

Fall 2004
 
 
Instructor &
Office Hours

Course Schedule,
Notes and Assignments

Textbook &
Reference Texts

Tutorials

Grading &
Classroom Policies

Links

Homepage

 

Grading:

Classroom Policies:

  • As per the Rowan College of Engineering policy, no eating or drinking in the classrooms and laboratories.
  • Lab projects must be worked in teams of not more than 2students. Each individual is  expected  submit a separate Lab Report (except for Final Project Report - see below). For report format, click here.

Final Project:

Requirements

As part of your final exam, you are required to design and develop an artificial neural network application of your choice; simulate the system and obtain results. Your network can perform typical functions such as signal/image classification and/or characterization (interpolation); although you are not limited to these. The development of new network architectures, measures of network performance and prediction confidence, etc. are also good projects. Discussion and implementation of current ANN literature (IEEE Transactions on Neural Networks, etc.) are also welcome.

Project teams will led by a Graduate student. A single final report from a team is acceptable.

Schedule

An e-mail (shreek@rowan.edu), Fax (856-256-5241) or a written/printed document presenting your project proposal must be sent to the instructor not later than Monday, Oct 4, 2004. This proposal must not exceed one half page.
Project deliverables include a  final project report (format follows) and a comprehensive presentation (~45 min + Questions) describing your work. The report is due by the end of the Fall semester Final’s week (Dec 20, 2004).

Report Format

The report should not exceed 15 double-spaced pages, 12 pt. font (including all figures and graphs, excluding appendix). The required final project report format is:
  1. Introduction – general description of the project topic/research paper.
  2. The problem statement – why is an artificial neural network required? What are the issues?
  3. Objective – what are you proposing to do/show?
  4. Approach – describe the method you have adopted.
  5. Implementation results – include your graphs, tables, etc. here.
  6. Conclusions – what have you learned? How can the algorithm be improved?
  7. Appendix – listing of the source code (exclude these pages from the total page count).