Introduction to Artificial Intelligence

CS471 • Fall 2010 • Time: TR 3:00 pm - 4:15 pm  • Location: LWSN B134

Schedule

Instructor

Professor Yuan (Alan) Qi
Lawson 2142L • alanqi[at]cs purdue edu
Office hours: Tues 4:15-5:15pm

Teaching assistant

Jyotishka Datta
Lawson 2149-3 • jdatta [at] purdue edu
Office hours:  Tues: 1:30-2:30pm  Wed: 4:00-5:00pm Fri: 1:30-2:30pm

Description

This course provides an introduction to foundational areas of artificial intelligence and current techniques for building intelligent systems. Tentative topics will include: problem solving, state-space representation, heuristic search techniques, knowledge representation, logical reasoning, reasoning under uncertainty, machine learning, and natural language processing.

Note that CS471 will be helpful for both undergraduate and graduate students who are interested in the graduate level course, CS 578 /STAT598N Machine learning, but lack background in machine learning and statistics.  The graduate level course will be available in 2011 spring.

Prerequisites

CS251: Data Structures.

Text

S. Russell and P. Norvig (2002). Artificial Intelligence: A Modern Approach. Prentice Hall, 2nd edition.

Assignments and exams

There will be six homework assignments, including three programming assignments. Programming assignments may be written in Java, Python, matlab, C or C++.

Homework assignments will be posted to Blackboard, and they should be submitted there, unless otherwise noted. In general, questions about the details of homework assignments should be directed to the TA, though you should feel free to mail the instructor whenever you have a question. Solutions and grading criteria will be available on Blackboard when homework is returned to students.

Exams will be closed book and closed notes. Review sessions will be held in the class before each exam. The exam questions will focus on conceptual understanding of the key ideas from the course, including problem definitions, algorithms, and data structures.

Grading

Grades will be posted on Blackboard. If you think a grading error was made on an assignment or test, or if you do not receive a homework assignment or exam back, you must talk to the TA or the instructor within a week of when it was returned.

Late policy

Assignments are to be submitted via Blackboard by the due date listed. Each person will be allowed five days of extensions which can be applied to any combination of assignments during the semester without penalty. After that a late penalty of 10% per day will be assigned. Use of a partial day will be counted as a full day. Assignments will not be accepted later without the express permission of the instructor. In general, additional extensions will be granted only due to serious and documented medical or family emergencies.

Academic honesty

Please read the departmental academic integrity policy. This will be followed unless I provide written documentation of exceptions. However, I encourage you to interact amongst yourselves: you may discuss and obtain help with basic concepts covered in lectures or the textbook, homework specification (but not solution), and program implementation (but not design). However, unless otherwise noted work turned in should reflect your own efforts and knowledge. Sharing or copying solutions is unacceptable and could result in failure. You are expected to take reasonable precautions to prevent others from using your work.

Course outline

Introduction
What is artificial intelligence? Overview of AI history and associated application areas.

Search
Problem solving as search, heuristic search, constraint satisfaction, adversarial search.

Reasoning with uncertain knowledge
Basic probability and statistical reasoning, Bayesian and Markov networks, exact and approximate inference methods.

Learning
Learning as search, parameter estimation, structure learning, overfitting, bias/variance.

Reasoning with logic
Propositional logic and first-order logic, logical reasoning and inference.

Reasoning with logic and uncertainty
Combining Bayesian and Markov networks with first-order logic, including representation, learning and inference algorithms.

Applications
Applications of machine learning and probabilistic inference, such as natural language processing.