CSE 730: Survey of Artificial Intelligence II: Advanced Topics

Course Description

A survey of advanced concepts, techniques, and applications of artificial intelligence, covering topics in machine perception (vision, audition, speech and language processing), reasoning (machine learning and inference), and machine production (speech synthesis).

Level, Credits, Class Time Distribution, Prerequisites

Level Credits Class Time Distribution Prerequisites
UG32clCSE 630

Quarters Offered, General Information, Exclusions, Cross-Listings, etc.

Objectives

Text:

Topics

Number of Hours Topic
2 Introduction, probability theory
6 Probabilistic inference
4 Machine learning
2 Computer vision
2 Computer audition
2 Automatic speech recognition
2 Spoken dialogue systems
3 Natural language processing
2 Abstract reasoning
2 Information retrieval
3 Exams, exam review

Representative Lab Assignments

Policy on Academic Misconduct

As with any class at this university, you are required to follow the Ohio State "Code of Student Conduct." If you are unfamiliar with this policy, you should read it at http://studentaffairs.osu.edu/resource_csc.asp. In particular, you should note that you are not allowed to, among other things, (a) knowingly provide or receive information during exams, (b) knowingly provide or receive assistance on homeworks unless I say it's OK, and (c) submit plagiarized (copied but unacknowledged) work for credit. If I suspect that any violation occurs, I am required to report the violation to the Council on Academic Misconduct. COAM will determine the guilt or innocence and appropriate penalties if any.

Guest Lectures

There will be a number of guest lectures (roughly 30 minutes each) this quarter that will outline some of the AI research pursued at OSU; you are responsible for material presented in these lectures. Treat the material in these lectures like any other part of the course.

Announcements

Announcements will be made via the course website. I will also monitor discussions on the cse.course.cse730 newsgroup and answer as appropriate, but I see this more as a forum for you all to discuss topics from class.

Grading Plan

This is the approximate weighting of the different components of this course:
Homeworks3x10=30%
Quizzes2x10=20%
Final Exam20%
Final Project25%
Participation5%

Homeworks: There will be four homework assignments, some written and some programming. The grade from the lowest homework assignment will be dropped; this means that you can miss one homework assignment with impunity. Homeworks are due at 11:59 PM on the day assigned. Late homeworks will be penalized 10 points for every hour late or fraction thereof. Submitting at 1:00 AM, for example, will be 20 points deduction, since it is one hour and one minute late. You will need to submit your homeworks via the ``submit'' command on the unix clusters. All code must be runnable on the unix system, even if you've developed it on other platforms. You may use the programming language of your choice.

Quizzes & Exams: Excuses from exams will only be given with prior notice (in the case of reasonable conflicts), or with a doctor's note. If you're sick enough to miss a class, then you should be at the health center; they can give you a note.

There will be two 1/2 period quizzes within the first six weeks of the quarter, in order to get you grading feedback as soon as possible. The final exam will be take-home, to be handed out the last Tuesday of class and due at the start of the last Thursday class. Please note the quiz/exam dates now (Oct 11, Oct 25, Nov 27-29).

Final project: You will be expected to apply machine learning techniques to an ``interesting'' problem; you are encouraged to propose a project based on your interests (grad students are particularly welcome to incorporate their own research), although this is not necessary (I will provide some sample projects). Proposals for projects must be cleared with me, and are due in the 6th week of the quarter. You are encouraged to make an appointment with me sometime early on to discuss potential projects. You must work in teams of two or three (larger groups == more substantial projects); exceptions to this rule need to be cleared with me. Each team will turn in a written report of 5-10 pages double spaced, along with supporting code, by the start of the final exam time slot. During the designated final exam period, we will have a poster session, where your team will give me a 5 minute summary of what they have accomplished.

Students enrolled in my 788 class this quarter may propose a joint project between the two classes, but it is expected in that case that the project be much more extensive, and have at least two different aspects so that one focuses on one aspect for 730 and the other for 788. See me for details.

Class participation: Participation is an important part of your educational experience in this class, so 5% of your grade will be determined by your participation.

Newsgroup / Mailing list

I have set up a newsgroup/mailing list reflector via the mailman interface. If you prefer to get your class news via email rather than netnews, subscribe to the mailing list. If the technology works as advertised, anything sent to the mailing list will end up on the newsgroup (cse.course.cse730) and vice-versa, however I have had problems with this gateway in the past. You may only post to the mailing list if you are a subscriber, otherwise you need to use the netnews interface. Subscribers to the mailing list need to be approved by me; I'm only allowing class members to subscribe.

Reminders