CSE 592 Applications of Artificial Intelligence Winter 2003 SLN Course Title Meetings Days Times Location 8635 CSE 592 TU APPS ARTIF INTEL Th 630-920pm SIG 322

Syllabus

Overview

This course will provide a broad overview of current research in artificial intelligence, with an emphasis on algorithms and representations that can be put to use in solving practical problems now or in the near future.  Topics that we will cover include:

• Planning and heuristic search
• Reasoning with uncertain information
• Machine learning, including decision trees, neural nets, datamining, and reinforcement learning
• Natural language understanding
• Autonomous systems

Coursework will consist of 4 assignments plus a significant project.

Textbooks

Please obtain copies of both textbooks by the start of class.

• Stuart Russell & Peter Norvig, Artificial Intelligence: A Modern Approach, 2nd Edition
  • I just received my own copy of the 2nd edition of the book (two days before class begins!) and discovered that the improvements over the first edition are much more extensive than I had expected.  Please buy a copy of this edition: they are in stock at the University Bookstore on University Avenue.  I apologize to anyone who already bought the first edition for the inconvenience.
• Tom Mitchell, Machine Learning

Meetings

Students are encouraged to meet with me to discuss course material, projects, or other questions about AI and computer science. Send me email a day or two in advance to set a time to talk in person or by phone. Because my schedule changes frequently, in your message please specify several different times at which you could meet, and I will confirm one for which I am available.   The TA, William Portnoy, is the best person to ask questions about assignments, grading, and the UW computing environment.

Remote Education

This course may be taken remotely from the Microsoft Redmond campus.  See http://www.cs.washington.edu/masters/dl_tech/ for details.  Other questions about the remote learning infrastructure should be sent to Fred Videon or Rodney Prieto. 

Course Mechanics

• There is no midterm or final.  Grades will be based on projects, written assignments, and class participation.
• Turning in assignments late creates an extra burden on the instructor and TA.  If you suspect that you will not be able to complete an assignment on time due to work or family commitments, please discuss the matter with me as soon as possible.  Unexcused late assignments will be penalized.
• This course may be taken remotely.  Fred Videon and Rodney Prieto make the magic work.
• If you took CSE 592 when it was taught by Pedro Domingos as a course on Datamining you may take the course again this quarter.  About 1/3 of the material we cover will overlap what was covered in the earlier version of the course.
• Programming projects may be done in any language on any computing platform.  You are encouraged to find and reuse code libraries you find on the web.  You must clearly document all such code reuse:  where you got it, what it does, how you modified it.  You should not, of course, simply hand in some project you found on the web as your own.  If you have any doubts about whether a case of code reuse is legitimate, please discuss it with me.
• Everyone is encouraged to use the course mailing list <cse592@cs.washington.edu> to discuss course material (you should be already subscribed using your @u.washington.edu email address; you can change your subscription at this web page).
• If you require academic accommodations, please contact Disabled Student Services, 448 Schmitz, 543-8924 (V/TTY). They will give you a letter requesting academic accommodations; please present the letter to me and we will make the accommodations that you need for class.