CS 175: Project in Artificial Intelligence
Fall 2007
Course Goals:
The main goal of this class is to have each student implement a working
artificial intelligence system. For this quarter we
will be primarily focusing on building a system which can analyze and
recognize
faces
in digital images. This is something which humans can easily do, but
which
is difficult for computers. The project will involve the implementation
of a variety of learning algorithms and image analysis
functions.
The overall goals of the class are to:
- Learn how to use the MATLAB scientific programming environment
- Understand the general principles of how computers can perform
simple
learning,
recognition and classification tasks
- Learn about automated image analysis and why it is a difficult
problem,
but how in some cases we can use learning algorithms to build
reasonably
accurate and useful image analysis systems
- Design, implement, and test a system (composed of component
algorithms)
which can learn to detect and/or classify face images
- Learn how to experimentally evaluate and test your system.
General Information:
- Class Times, Location, and Schedule:
- Location: Multipurpose Science
& Technology Building, MSTB
226: Lab B (Room 226)
- Lecture Time: Tuesdays and Thursdays, 9:30 to 10:50am
- Course code = 34360
- Lab Hours:
- CS 364 lab, machines with MATLAB (see MATLAB Resources), any hours that CS 364 is open.
- Multipurpose Science
& Technology Labs, MSTB
226: Lab A and B (Room 226), any time the labs are open and not booked for another class
- Weekly
Schedule, Lecture Notes, and Assignments
- MATLAB
resources (general information on MATLAB, where to find it, etc)
- Who to Contact with Questions
- TA: Nathan Sutter
- email address: nsutter@uci.edu
- Office Hours: 4 to 6pm, Mondays, CS 364 Lab
- additional information on Nathan's TA Web page
- Instructor: Professor
Padhraic Smyth
- email address: smyth@ics.uci.edu
- Office Hours: Thursdays, 3:30 to 5pm, BH 4212
- Useful Reference Texts (these are not required!)
Grading
Your grade will be determined based on the following factors:
- Completion of project milestones on time (design, MATLAB coding,
testing)
- Correctness and quality of your work, i.e.,
- does your code function correctly,
- how well the code is designed/written/tested,
- how well it is documented,
- Understanding of basic concepts underlying the algorithms
- Your ability to evaluate your results and the presentation
quality of
your
project reports.
There will be two parts to the class:
- The first 5 weeks or so: weekly assignments using MATLAB: worth approximately 40% of your grade, with the lowest scoring assignment dropped.
- The second 5 weeks: project proposals, project progress reports, project presentation in class, and final project report: worth approximately 60% of your grade
Academic Honesty
Academic honesty is taken very seriously. You are allowed to verbally
discuss
the design and implementation of your code with other class members,
but
under no circumstances can you copy *any* code or written material from
anyone or anywhere. All code and written material submitted must be
code
or written material you have personally written during this quarter,
except
for any library or utility functions which we supply. Failure to adhere
to this policy can result in a student receiving a failing grade in the
class. It is the responsibility of each student to be familiar with
UCI's
current academic honesty policies. Please take the time to read the
current
UCI Senate Academic Honesty Policies.
Face Recognition Resources on the Internet