CS 430 - Artificial Intelligence

Spring 2005 - Guidelines for AI Research Paper

Out: March 3, 2005
Topic Choice Due: Thursday, March 17, 2005
Outline Due: Thursday, March 31, 2005
Draft Due: Thursday, April 7, 2005
Paper Due: Thursday, April 14, 2005

The main goal of this assignment is to research a topic in Artificial Intelligence that will not be covered in any detail in the class. Each student will write a report and give a presentation on a topic of their choosing.

Logistics

Each student is responsible for:

• an approximately 10-page written report, due on April 14. The report will be distributed to the rest of the class at the time of the presentation
• a 30-minute class presentation
• 3-5 questions suitable for the take-home final exam on the material covered in the presentation and in the report, due one class period after the presentation.

Paper topics must be approved by the instructor no later than Thursday, March 17. Some possible topics are listed below. An outline of the paper with references must be submitted no later than Thursday, March 31. A draft of the paper is due no later than Thursday, April 7. Submissions earlier than these deadlines would be appreciated. The final paper is due Thursday, April 15.

Presentations will be given on April 19, 21, and 26. Order of presentations will be assigned by the instructor after topics have been chosen.

Report Content

Your report should follow standard formatting for technical reports. This includes a cover page, an introduction/overview, a background section, as many sections needed to cover the relevant topics, a conclusion, and a numbered list of references in alphabetical order by first author. Citations should be of the form ``[#]'' where # is the number of the reference in the list. The report pages should be numbered starting with the cover page (but the cover page should not have a number printed on it). Your report should contain information on the following:

1. An overview of the research topic
2. Problem statement giving the goal of the research. What problem it being solved? Or what benefits are expected to be realized?
3. Background necessary to understand the research topic. This may include definitions, reminders of algorithms previously studied, etc.
4. The bulk of the report should be on explaining the research topic and showing how it is used.
5. A conclusion tying the research topic to current applications.

Presentation

Each student will present one 30-minute presentation on their research topic, including time for questions. The presentation should be an overview of the report pointing out the highlights. Please do not read your entire report to the class. Throughout the lecture, the presenter should be prepared to answer questions.

Presentation software may be used, but is not required. Overhead slides may be used. Instructor can have slides made from handouts.

Research Topics and Resources

Here is a list of possible research topics. This is certainly not an exhaustive list. You may suggest others, but get instructor approval FIRST.

1. Definition and description of work in traditional AI application areas not covered in class such as expert systems, natural language processing, machine learning, computer vision, etc.
2. Definition and application of AI techniques not covered in class such as genetic algorithms/programming, fuzzy logic, neural networks, higher-order constraint propagation, Bayesian networks, hidden Markov models, etc.
3. History of and state-of-the-art application of AI techniques to real-world application areas such as constraint propagation in printed circuit routing or scheduling, A* search and other path-finding algorithms for computer game non-player characters, adversarial search in turn-based game-playing (e.g., chess, checkers, bridge, backgammon, or Scrabble), etc.