CIS 521 - Fundamentals of AI - Fall 07

Lectures: Towne 303, Tuesday and Thursday, 1:30-3pm

Instructor: Ben Taskar
Email: taskar@cis
Office hours: North Levine 464, Thursday, 3-5pm

Teaching Assistant: Ben Sapp
Email: bensapp@seas
Office hours: Levine 612, Tuesday, 3-5pm

Administrative Assistant: Cheryl Hickey
Email: cherylh@cis
Office: 502 Levine

Announcements

We will use Blackboard (https://courseweb.library.upenn.edu) for lecture notes, assignments and other materials and communications. Please send email to cis521@seas if you're having trouble accessing Blackboard.

Course description

Modern AI uses a collection of techniques from a number of fields in the design of intelligent systems: probability, statistics, logic, operations research, optimal control and economics, to name a few. This course covers basic modeling and algorithmic tools from these fields underlying current research and highlights their applications in computer vision, robotics, and natural language processing.

Intended Audience

This course is targeted at advanced undergraduates (juniors and seniors) and graduate students, both Masters and PhD level, providing a rigorous introduction to modern AI techniques and principles.

Comparison to CSE 391 - Artificial Intelligence
CSE 391 is targeted at undergraduates and tends to assume much less analytical and programming background. There will be a large overlap in the topics, but the material will be presented at a more challenging and rigorous level. The goal is to make CIS 521 the "honors" version of 391, not a sequel to it.

Comparison to CIS 520 - (formerly "Artificial Intelligence and") Machine Learning
CIS 520 is targeted at graduate level students. Its traditional focus (for several years) has been machine learning, to the exclusion of other fundamentals of AI. There will be 20-30% overlap in the material. (CIS 520 has been renamed Machine Learning.)

Part I: Agents, Search and Planning
Motion Planning, Games
Optimization, Heuristic and Randomized Search
Logic, Satisfiability, CSPs, Action Calculus
Part II: Probabilistic Reasoning and Learning
Supervised and Unsupervised Learning Basics: Linear Classifiers and Clustering
Hidden Markov Models, Graphical Models (Bayes and Markov Nets), Decision Diagrams
Temporal Reasoning: Tracking and Filtering
Reinforcement Learning Basics: Markov Decision Processes, Q-Learning
Part III: Connections and Applications
Topics in Vision, Natural Language Processing, Speech
Philosophical Questions

Materials

Required Text: S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, Second Edition,
Prentice Hall 2003, ISBN: 0-13-790395-2.

Selected readings from other books and papers will be distributed as electronic or hard copies.

Pre-requisites

Students are expected to have the following background:

Evaluation

The midterm will be an open-book, open notes exam, which will encompass material covered in the lectures and assigned in the readings.

Problem Sets and Programming Assignments

There will be two problem sets and two programming assignments. Programming assignments will require programming the interesting parts of the problems. We will provide the necessary software infrastructure for the programming assignments in Python.

Students will be allowed to work together on written homeworks. Students may discuss the homework to understand the problem and reach a solution. However, each student must write down the solution independently, and without referring to written notes from the joint session. In other words, each student must understand the solution well enough in order to reconstruct it by him/herself. In addition, each student must write on the problem set the set of people with whom s/he collaborated. On the programming assignments and on the final project competition, you can work in groups of up to two students. For these, each group should submit one solution set.

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.

In addition to homeworks, we will hand out exercises. The students are expected to solve the exercises, but not hand them in. The exercises will not be graded. However, we strongly recommend that you try to solve the exercises, as it will greatly enhance the benefit that you will get from attending the sections (where the exercises will be solved).

Important note on the honor code: The purpose of problem sets in this class is to help you think about the material, not just give us the right answers. You are free to use online resources for learning more about the material covered in class; however, you should not look online for solutions to questions in the problem sets.

Project

Students are required to complete a significant project. Students will be given an open-ended challenge problem, set up as a competition. Details on the challenge problems will be announced in class. Your solution will be judged both in terms of its performance (at the final project competition held at the end of the semester), and in terms of the quality and novelty of your ideas (as described in your writeup).

For each of the two challenge problems, at the final project competition we will declare a winning and a runner-up team. Each member of the winning team will receive 3% extra credit; each member of the runner-up team will receive 1.5% extra credit. You will also submit a final writeup describing your solution to the challenge problem.

Late homeworks: Recognizing that students may face unusual circumstances and require some flexibility in the course of the quarter, each student will have a total of seven (+ 2 = 9 total) free late (calendar) days to use as s/he sees fit. Once these late days are exhausted, any homework turned in late will be penalized 20% per late day. However, no homework will be accepted more than four days after its due date. Each 24 hours or part thereof that a homework is late uses up one full late day. Late days are not permitted for the final project code or the final project writeup. For work done in teams, late days apply individually to all of the team members. Late homeworks should be handed in as follows: Write the date and time on the assignment and then drop it off at Cheryl Hickey's office (Levine 502). It is an honor code violation to write down the wrong time.

Extra credit will be given to the top performers in the project challenge problem. It may also be given to some projects that go substantially above and beyond the requirements in finding creative solutions to the task. In addition, we will use class participation as a factor in determining the final grade in borderline cases, so we encourage you to attend class and participate actively.