Can a computer be taught to read words aloud, recognize faces, perform a medical diagnosis, drive a car, play a good game of backgammon, balance a pole, or predict physical phenomena? The answer to all these is yes. All these applications and others have been demonstrated using varieties of artificial neural networks. In this course we will explore neural network models from a theoretical and practical point of view, paying close attention to their connections to cognitive science, neuroscience, and biological nervous systems. Topics to be discussed include supervised and unsupervised learning algorithms, feedforward and recurrent network architectures, and evolutionary approaches to neural network design.
Prerequisites: CS 60 and Linear Algebra, or permission of instructor.
Prof. Jim Marshall
Office: Andrew Science Building 261, Pomona College
Office Hours: Mondays 2:30-3:30pm, Tuesdays 2:30-3:30pm, Thursdays 11:00am-noon, and by appointment.
Phone/Voicemail: (909) 607-8650 (extension 78650 on campus)
Mondays and Wednesdays 1:15-2:30pm, Millikan 218 (Pomona)
All students enrolled in CS 152 need a Unix account on turing. If you do not have a turing account, you need to fill out this form, contact the HMC CS staff, and read the CS System Policies.
Introduction to Neural Networks
by James A. Anderson
You are expected to attend all lectures. There will be some homework and programming assignments, but no exams. A substantial part of your grade will be from a final project involving either the creation of a working neural network application or a research paper. The grade on the project will be determined by the comprehensiveness and degree to which you explored competing approaches. The projects will be presented orally. We will also read a number of papers from the neural network literature. At some point during the semester, each student will give a short presentation to the class on a paper from the literature.
Your final grade will be determined as follows:
|35%||Homework and Programming Assignments|
|15%||Paper Presentation||10%||Class Attendance and Participation|
Late Homework Policy
You are strongly encouraged to come to my office hours whenever you are having difficulty with the material. If you are confused about something, don't stay that way! Staying confused will only make things worse later. Come see me as soon as possible so that we can clear up the problem.
If you want to see me but can't make it to my office hours, I'll be more than happy to schedule an appointment. Ask me about it in class, send me some e-mail, or leave a message on my voice mail. You can also try to catch me outside of my regular office hours, though I can't guarantee that I'll always have time to meet with you right then.
The highest level of academic integrity is expected of every student. You are strongly encouraged to discuss ideas and approaches to problems with your classmates on a general level, but unless I tell you otherwise, the work you hand in must be exclusively your own. Effective learning is compromised when this is not the case.
Accordingly, you should never read or transcribe another student's code or solutions, exchange computer files, or share your code or solutions with anyone else in the class. Under no circumstances may you hand in work done by, or in collaboration with, someone else under your own name, with the exception that you may freely use any code that I explicitly provide to you.
When in doubt, credit the people or sources from whom you got help. This also goes for any help obtained via the Internet. You will not lose any points for acknowledging significant help obtained in a legitimate fashion. If you are ever unsure about what constitutes acceptable collaboration, just ask.
Failure to abide by these rules is considered plagiarism, and will result in severe penalties, including possible failure in the course. Please do not put me, yourself, or anyone else in this unpleasant situation.