This course covers various aspects of extending AI theories and techniques from the single-agent case to the multi-agent (MA) case. The topics include: Modal logics of knowledge and belief, logics of belief change, multi-agent probability systems, introduction to non-cooperative game theory, mechanism design and auctions, cooperative and non-cooperative communication, and multi-agent learning. Emphasis is on representation techniques and algorithms, drawn primarily from two disciplines - computer science (and, through it, philosophy) and game theory.
The goal is to give an informative and balanced introduction to each area, relate them to one another when possible, and in each case dig sufficiently deeply to tackle challenging technical material. To accomplish this, students will be expected to do a fair amount of reading on their own.
Students are expected to have a basic knowledge of probability theory, as well as knowledge of basic computer science concepts, including data structures, algorithms and complexity. The course is appropriate for advanced undergraduates, theoretically minded Masters students, and PhD students interested in the area.
These books are not required, but are good references and excellent places to learn more. They have all been requested to be put on reserve in the Math & CS library.
The course will have open-book midterm, a final paper (details
TBA), and four homework assignments. Students will be responsible for material covered in the lectures and assigned in the readings. For the schedule, we will work roughly from the Homeworks are individual assignments (as is the midterm, of course). It is quite acceptable to consult with others to clarify the course material, ask questions, or brainstorm ideas.
However, the solutions and write-ups must be your own. Please also note that some reuse of problems from previous years is possible, so referencing prior years' solution sets when solving the current homework is not allowed. We try very hard to make questions unambiguous, but some ambiguities may remain. Ask if confused or state your assumptions explicitly.
Reasonable assumptions will be accepted in case of ambiguous questions. If necessary, changes/corrections to the homework will be posted to the mailing
list. To turn in homework, please give us a hard-copy by the end of class on the
due date.
A grading penalty will be applied to late homeworks. (Note that the final paper cannot be turned in late.) Recognizing that students may face unusual circumstances and require some flexibility in the course of the quarter, each student will have a total of two free late
(calendar) days to use as they see fit. Once these late days are exhausted, any homework turned in late will be penalized at the rate of 25 points (out of 100) per late day. Under no circumstances will a homework be accepted more than three days after its due date. Late days are from
2:05pm to 2:05pm (i.e. not coming to class to turn in the homework = late). Late homeworks should be handed in to one of the course staff. If none are
available (e.g., on weekends), write the date and time on the assignment and e-mail the TA for instructions. It is an honor code violation to write down the wrong time.
Please do not e-mail grading questions to the teaching staff. If you want the TA to explain why he took points off, you can talk to him after class or during office hours. If you want a regrade, please write an explanation and give the homework and the explanation to the TA during office hours or after class.
Homework Details
Late Days
Regrades
Home