Artificial Intelligence (CS 436)

(Updated: 12/18/08)

Course Description

From the catalog: The fundamental bases of artificial intelligence: knowledge representation, search, and learning. Applications include game playing, neural networks, and expert systems. Common Lisp and CLOS are introduced.

This course examines the concepts, techniques, applications, and theories of Artificial Intelligence. Given the broad range of topics addressed by the AI field, topics for discussion must, necessarily, be limited. The course is organized around two main themes--symbolic AI and numerical AI--and these are covered in the first and second halves of the course, respectively. The topics in symbolic AI will emphasize logic and search. The topics in numerical AI will emphasize probabilistic methods, graphical models, and statistical learning. These topics are bracketed by discussions on the definition of AI and philosophical issues surrounding AI research

Course Dates

Wednesday, September 3, 2008 through Friday, December 19, 2008

Time and Location

EPS 108, MWF, 12:00-12:50 pm

Prerequisites

CS 223 Data Structures and Algorithms

It is also a good idea to have experience with Discrete Mathematics and Probability. You should also have completed at least two semesters of Calculus.

Professor Information

Dr. John W. Sheppard
EPS 363
Phone: 994-4835
Email: john dot sheppard at cs dot montana dot edu

Office Hours: MWF 11:00-11:45 am

Course Objectives

At the end of this course, the student will be able to:

• Formulate and assess problems in artificial intelligence.
• Assess the strengths and weaknesses of several methods for representing knowledge.
• Assess the strengths and weaknesses of several AI algorithms in areas such as heuristic search, game search, logical inference, statistical inference, decision theory, planning, machine learning, neural networks, and natural language processing.
• Implement software solutions to a wide-variety of problems generally considered to require artificial intelligence.

Course Readings

• Artificial Intelligence: A Modern Approach, Stuart Russell & Peter Norvig, Prentice-Hall, 2nd Edition, 0-13-790395-2
• Relevant research papers as required.

Course Evaluation

Grading will be based on four homework assignments, each of equal weight (10% each), three programming assignments, each of equal weight (10% each), class attendance (10%), and a comprehensive final examination (20%). The homework assignments will focus on clarifying and implementing ideas presented in class or in the readings. Programming assignments will be split into two parts. The first part will focus on implementation and the second part will focus on reporting your results. The weights of these two parts will vary by assignment. The student may use any language and machine for satisfying the programming requirements. The student may be expected to write one or two short papers consisting of brief reviews and critiques of current research papers in Artificial Intelligence.

Here is a summary of actual assignment point values:

• Assignment 1 (100 pts)
• Assignment 2 (100 pts)
• Assignment 3 (100 pts)
• Assignment 4 (100 pts)
• Program 1:
• Programming (70 pts)
• Writing (30 pts)
• Program 2:
• Programming (50 pts)
• Writing (50 pts)
• Program 3:
• Programming (30 pts)
• Writing (70 pts)

Class Schedule

Homework (For solutions, email instructor for username/password)

Note that all due dates for homework have been updated.

• Homework 1: Due Wednesday, September 24 (solutions)
• Homework 2: Due Monday, October 27 (solutions, z-test spreadsheet)
• Homework 3: Due Monday, November 24 (solutions)
• Homework 4: Due Monday, December 15 (by 12:00 noon, in my office) (solutions)

Programming

• Program 1: Due Monday, October 20
• Program 2: Due Friday, November 21
• Assignment
• Paper Template
• Program 3: Due Monday, December 15 (emailed to john dot sheppard at cs dot montana dot edu by 12:00 noon). Note: Students will be giving the option of making this a group assignment. Students may pair up to complete a joint program but write separate reports.
• Assignment
• L Track
• O Track
• R Track

Final Examination

• Date: December 18
• Time: 8:00-9:50 am
• Location: EPS 108
• Solutions

Policy on Academic Misconduct

Academic misconduct is unacceptable. It is the responsibility of all full-time, part-time or non-degree (special) students to adhere to strict standards of integrity in their professional and scholarly activities, as well as to high standards of conduct in their non-academic activities. Misconduct will be treated swiftly and harshly.

Examples of academic misconduct:

• Cheating on Examinations
• Use of unauthorized materials during examinations and in completing assignments.
• Consultation of unauthorized materials while being excused from an examination room.
• Discussion of an exam's contents during its administration.
• Copying answers from another student on an examination.
• Obtaining an examination or answers to an examination prior to its administration
• Studying from an old exam whose circulation was prohibited by the instructor.
• Plagiarism
• Submission of the same or substantially similar work of another person, even if re-worded
• Use of another person's work while representing it as your own
• Improper documentation of quotations, ideas, or paraphrased passages taken from published or unpublished works
• Reuse of Assignments
• Submission of the same or substantially similar assignment to fulfill the requirements of more than one course.
• Improper Use of the Internet
• Plagiarism from a published or unpublished Internet source.
• Improper documentation of an Internet source.
• Use of paper writing services of paper databases on the Internet.
• Improper Use of Electronic Devices
• Consultation of unauthorized electronic devices during an examination.
• Use of electronic devices to communicate within or outside of an examination room.
• Storage of test answers, class notes, and other references in electronic devices during examinations.
• Unauthorized Collaboration
• Collaboration when solving homework problems or writing lab reports, computer programs, or papers, unless explicitly approved by the professor.
• Alteration of Graded Assignments
• Submission of an examination or assignment for a re-grade after making changes to the original answers or text.
• Forgery or Falsification
• Falsification or invention of data in a laboratory experiment.
• Citation of nonexistent sources or creation of false information in a written assignment.
• Attributing to a source ideas or information that is not included in the source.
• Forgery of university documents.
• Impersonating a faculty member.
• Lying
• Request for special consideration from professors or university officials based upon false information or deception.
• Fabrication of a medical or emergency excuse.
• Claiming falsely to have completed and/or turned in an assignment.
• Falsely reporting an ethics violation by another student.
• Facilitating Academic Dishonesty
• Intentionally or knowingly aiding another student to commit a violation of academic conduct.
• Allowing another student to copy from one's own examination during its administration.
• Providing copies of course materials whose circulation was prohibited to students enrolled in or planning to take that course.
• Taking an examination or completing an assignment for another student, or permitting another student to do so on one's behalf.
• Unfair Competition
• Willfully damaging the academic efforts of other students.
• Stealing another student's academic materials.
• Denying another student needed resources.

Policy on Assignments

This course has several assignments requiring outside work of the students. The assignments are critical for gaining understanding and experience using the materials presented in class. Due to the importance of these assignments, the following policy is set forth.

1. All assignments will be completed by the individual student and will be the original work of that student.
2. All assignments are due at the beginning of class on the dates indicated in the syllabus. No assignments will be accepted late without prior approval of the instructor (other than exceptions noted below). Approval will not be granted based on personal time-management issues.
3. Unapproved late assignments will receive no credit. Approved late assignments may still receive a penalty, depending on the circumstances. It is the responsibility of the student to ensure that the instructor is kept informed of any problems related to turning in assignments on time. Only serious, uncontrollable circumstances (such as serious illness or family tragedy) will result in accepting late assignments without prior notification. In such cases, documentation of these circumstances must be provided.
4. Attending class sessions is critical to a successful course. If an absence is anticipated, please notify the instructor beforehand by phone or email. Unexplained absences will adversely affect the final grade.
5. All written assignments are expected to be typed or neatly printed. Avoid hand drawn figures if at all possible. If the assignments are not legible, they will be returned to the student with a grade of zero. Be sure each assignment includes name, department, daytime phone number, and email address.
6. While a computer account is provided to all students, any language and any computer system can be used to complete the programming assignments unless otherwise specified in the assignment itself. That said, it is expected that all programs can be run on the department network, and students may be requested to demonstrate this.
7. All programming assignments must include fully commented code and several sample runs to demonstrate proper functioning of the assigned program. It is the responsibility of the student to ensure that the code is readable and understandable. It is also the responsibility of the student to ensure that the output is understandable and accurately reflects the functioning of the program.
8. The world-wide web provides a tremendous resource to both students and instructors, and students may be tempted to look for problem solutions on the web. Any student turning in an assignment with a solution obtained from the web must give full attribution to the source of that solution. Failure to do so is plagiarism and is grounds for failing the course. Even with proper attribution, credit received for the assignment will depend on the nature of the web information used and on the problem assigned. (See policy on web usage.)

Policy on Web Usage

The world-wide web provides a resource for finding and using a tremendous amount of information in the computer science and engineering fields. As computer scientists, we are able to use the web to maximize our productivity in all aspects of our life, including home, work, and school, and in this class, all students are encouraged to use the web as an educational resource. Unfortunately, as with any resource, use of the web can be abused to the point where the educational experience is diminished. As an attempt to limit such abuse, the following policy on using the web in this class is set forth.

1. Students are free to explore the web to visit sites related to topics discussed in this course insofar as such exploration identifies material that elucidates and expands on material related to or discussed in class.
2. Students are not permitted to seek out solutions to any of the problems assigned unless the assignment specifically states that web use is permitted.
3. On programming assignments, students may not download solutions from a web site implementing similar or equivalent programs. Further, students are discouraged from even examining such solutions. Turning in a program obtained from the web will result in no credit for that program. Further, turning in a program obtained from the web without attribution to the web source constitutes plagiarism and will result in failure of the course.
4. On problem assignments, students may not download or examine solutions from a web site focusing on similar or equivalent problems. Turning in a solution from the web will result in no credit for that problem. Further, turning in a solution obtained from the web without attribution to the web source constitutes plagiarism and will result in failure of the course.
5. If there are any questions or even a hint of doubt concerning appropriate use of the web for completing class assignments, the student should consult the instructor for guidance.

Policy on Class Attendance

A large amount of material will be covered in a relatively limited amount of time. In addition, a fairly large amount of work will be done by the student. Consequently, class attendance is required. If a student must miss class for any reason, he or she should notify the instructor as soon as the absence is known. In the event of emergency absences, the instructor reserves the right to request an excuse from some cognizant authority such as a supervisor or physician. Note that class attendance accounts for 10% of the student's final grade.

Policy on Personal Communications Devices

It is unfortunate, but the advances in personal communications technologies has also resulted in the need for a policy concerning the use of these devices. Since students receiving and/or responding to pages or cell phone calls creates a distraction to other students, no pagers or cell phones will be permitted to be brought into the classroom without prior authorization of the instructor.