 Middle East Technical University Ankara, Turkey |
CENG569 Neurocomputing Spring 2005-2006 |
 Department of Computer Engineering |
| Information |
Lectures |
Policies |
Information |
||||||||||||||||||||||||||||||||||||||||
|
Announcements
|
|||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||
Policies
- Registration: The
size
of the class is limited and therefore, if necessary, I will try to make
the
class available to the most interested students.
- Objective: This course aims to familiarize the student with the neural network literature.
- Prerequisites: Consent of the instructor is
essential.
Calculus I and II (or equivalent level of math), Introduction to
Programming (CENG 210-220-230) and a Data Structures course.
- Grading:
- Class participation
(10%)
- 4 Projects (60%)
- Expected project report quality: Check the following reports as good examples.
- Paper presentation (10%)
- If you have done a specific project, then you will
present
your work.
- Otherwise, you will be asked to review a set of papers
and
make a 15 minute presentation in class.
- Homeworks (20%)
- All lectures have reading assignments, consisting of some chapters from the texbook as well as extra reading material (mostly papers). Related to these readings, you will be given a short list of questions to be answered in a prior lecture. The homeworks are expected to be submitted by the beginning of each class. Late submissions will not accepted. If you are not able to read the readings, then do not attempt to cheat by copying it from another one's homework.
-
Late assignments
- All assignments are due the start of a class. Assignments submitted after 15 minutes of a class are considered late.
- Homeworks submitted late will not be accepted!
- Project reports submitted late are
eligible for a maximum of 80% credit. Projects
submitted after 2 weeks of their due date are worth at most 60% credit.
Any project not submitted within two weeks of its due date receive no
credit.
- Paper presentations have to be made at the scheduled time. Re-scheduling [at least one week in advance] without legal (e.g. health) excuses will cause the loss of 50% of the credit.
- Presentation guidelines: You are given one or two papers to be presented. Your task is to prepare a 20 minute presentation about the paper[s]. You should fully understand the papers, and present it. The style of presentation is very important. Your slides should contain the main points that you will be talking [no full sentences!]. Try to make your slides "light" but informative. Resort to images/graphics wherever possible. Also rehearse your talk couple of times before your presentation. During and after the presentation you will be grilled by questions, and your performance in replying is also very important
- Additional material:
| [AR]
Neurocomputing: Foundations of Research Edited by J. A. Anderson, E.Rosenfeld, MIT Press, Cambridge, 1988. |
 |
| [DHS] Pattern
Classification R. O. Duda, P. E. Hart, and H. G. Stork, Wiley-Interscience, New York, NY, 2 ed., 2000. Available at the Reserve section of the library. |
 |
| [K]
Self-organization and associative
memory by T. Kohonen, Springer-Verlag, 1988. Available at the Reserve section of the library |
 |
| [MR] Parallel
Distributed Processing I and
II, by J. McClelland and D. Rumelhart . MIT Press, Cambridge, MA, 1986. |
  |
| [G] Studies of
the Mind and Brain, S. Grossberg, Reidel Press, Drodrecht, Holland, 1982. |
|
| [CG] Pattern
Recognition by
Self-Organizing Neural
Networks, Edited by G.A. Carpenter, and S. Grossberg, Cambridge, MA, MIT Press, 1994. |
 |
| [KSJ] Principles
of Neural
Science, E.R, Kandel, J.H. Schwartz, and T.M. Jessel, Appleton &Lange, 1991. |
 |
| Also other complementary articles that will be made available |