Arti cial Intelligence (AI) seeks to understand the mechanisms underlying thought and intelligent behavior, with a particular focus on their embodiment in machines. Core topics include the integrating perspective of intelligent agents plus how such systems can engage in: search and problem solving; symbolic and probabilistic knowledge representation and reasoning; planning; and machine learning. The course introduces both basic concepts and algorithms, and explores how to apply these in the construction of systems that can interact intelligently with complex environments.
The course is intended for undergraduate students in computer science or closely related disciplines, usually in their junior year. Graduate students should take CS561 rather than CS360L.
The courses CSCI 104L ("Data Structures and Object-Oriented Design") and CSCI 170 ("Discrete Methods in Computer Science") are necessary prerequisites. Overall, the prerequisites of CS360L include a solid understanding of data structures, algorithms and programming since you will have to be able to understand algorithms and read pseudo code. You should also know the basics of probability theory. Finally, you should know how to program in C/C++ since the projects will use these programming languages. Do not take this class if you cannot program. The most important prerequisite of all, however, is your interest in the class, motivation, and commitment to learning. If you are not sure whether this class is for you, come and talk to us.
Most readings will be chosen from the (required) textbook, which will be made available at the USC bookstore and is also readily available from many standard online retailers:
Stuart Russell and Peter Norvig Artificial Intelligence: A Modern Approach (most recent edition - currently: 3rd edition), Pearson/Prentice Hall (ISBN 978-0-13-604259-4)
The authors made extensive revisions from one edition to the next one. We therefore suggest that you buy the latest edition. Definitely do not use the first edition. We will not cover all of the chapters and, from time to time, cover topics not contained in the book.
Additional material will be provided as necessary.
The class web pages are maintained at http://idm-lab.org/wiki/360/. The discussion forum and scores are maintained at blackboard.usc.edu.
We will ask you to fill out two questionnaires: one at the beginning of class because we would like to learn a bit more about you so that we can tailor the class toward your skills and expectations; and one before the final so that we can improve future classes. Filling out the questionnaires is voluntary and anonymous. We hope that you will make use of this opportunity and choose to provide us with feedback.
We will try to make the lecture slides available on the class web pages before the lectures. The lectures are meant to summarize the readings and stress the important points. Thus, we expect you to read the corresponding part of the textbook before the lectures. If you miss a class, it is your responsibility to find out what we discussed in class, including which announcements we made in class. If there is something that you don't understand, feel free to interrupt the lecture or session with questions. Your active participation in class is crucial in making the class successful. Use your colleagues as a resource (they are working towards the same goal as you are), for example, by forming study groups or posting questions on the discussion forum that the TA monitors on a daily basis. If you need additional help, please feel free to go by the TA during his office hours. If lots of students are confused, the TA will give help sessions with additional examples.
To help you prepare for the exams, we will post "text-book style" homework assignments with short questions. We will not collect or grade your solutions. However, solving these assignments on your own is important as it will ensure your understanding of the material in preparation for the exams. The TA will hold a regular exercise session where solutions to some of these exercises are discussed. The solutions to each assignment will be posted on Friday of the week of the corresponding exercise session.
There will be three graded two-week projects. Most or all of the projects will involve both some programming problems in C/C++ and some theoretical problems. All of them have to be done individually. You are required to cite all resources you relied on for coming up with your answers. This includes people, web pages, publications and other write-ups. You are not allowed to use code or code snippets of others, that is, that you did not write yourself. You are not allowed to discuss with others how to solve the projects.
Please start to work on your projects early and hand them in early. There is a linearly increasing penalty for late submissions so that submissions that are 48 hours (or more) late receive no credit.
There will be two midterms and one final. The midterms will be written on Wed, Feb 25 and Wed, Apr 8 in class, and the final will be written during the officially scheduled date, which is Fri, May 8 from 2:00pm to 4:00pm according to http://classes.usc.edu/term-20151/finals/ (in our regular classroom unless announced otherwise). No makeups will be given. The exams will be open textbook ("Artificial Intelligence - A Modern Approach" only) and open printed or hand-written (but not electronic) notes (including, but not limited to, all material posted on this wiki). All exams will be comprehensive. Bring a calculator to all exams. Using computers, cell phones or similar equipment is not allowed.
We will grade on a curve by adjusting the maximum score of each project and exam individually so that the average percentage on that project or exam is 77.5%. Projects and exams have the following weights:
Project 1: 10%
Project 2: 10%
Project 3: 10%
Midterm 1: 20%
Midterm 2: 20%
Final: 30%
The intended grading scale is as follows.
90% and larger: A
85% - 90%: A-
80% - 85%: B+
75% - 80%: B
70% - 75%: B-
65% - 70%: C+
60% - 65%: C
55% - 60%: C-
50% - 55%: D+
45% - 50%: D
40% - 45%: D-
00% - 40%: F
The instructor reserves the right to adjust the grading method, including curving method and grading scale.
The instructor will assign grades from A to F, if warranted. There will always be some students who are very close to grade boundaries. There is nothing that we will do about that. Grades are based on performance, not need or personal circumstances, and the instructor does not negotiate grades. Thus, do not take CS360L (or take it completely at your own risk) if you need a certain grade, for example, because you are graduating or because you have been conditionally admitted.
CS360 is very exam-heavy. To receive a good grade, you will therefore need to perform well in exams. Please check the correctness of the grading and the posted scores immediately after we announce the availability of the scores. You will need to let us know about any grading issue with an exam, project or similar within 14 days of us posting the score for that exam or project. After that time, we will no longer entertain your requests for changes to your score. If you have a grading issue, you will need to discuss the issue first with the TA. If you cannot reach consensus, you can then appeal the grading issue to the instructor. Both the TA and the instructor might check the exam or project completely for grading issues and adjust your score up or down as appropriate.
We will not accept excuses unless you provide us with a note from a doctor (or similar professional) that verifies the problem and you told us about the issue IMMEDIATELY WHEN IT AROSE (not: after it has already affected your performance in class). We accept only true emergencies as excuses, such as your sickness or a death in your immediate family. We are sorry that we cannot make exceptions to these rules. So, please do not ask for them.
USC seeks to maintain an optimal learning environment. General principles of academic honesty include the concept of respect for the intellectual property of others, the expectation that individual work will be submitted unless otherwise allowed by an instructor, and the obligations both to protect one's own academic work from misuse by others as well as to avoid using another's work as one's own.
All students are expected to understand and abide by these principles. Scampus, the Student Guidebook, contains the Student Conduct Code in Section 11.00, while the recommended sanctions are located in Appendix A: http://scampus.usc.edu/university-student-conduct-code/. We will strictly enforce the Student Conduct Code and refer students to the Office of Student Judicial Affairs and Community Standards for further review, should there be any suspicion of academic dishonesty, and suggest that they follow the sanctions in Appendix A, should they find that there was academic dishonesty.
If you think you might need a recommendation letter at some point in time, please read Sven's FAQ.
At some point, you will have questions. For example, you might not be able to get code to run that we provided, there is something in the textbook that you do not understand, and so on. In this case, we encourage you to post the question on the discussion forums on blackboard.usc.edu and see whether someone can help you. If this approach does not generate the desired result, then the TA will be happy to help you in person. He does answer email but, unfortunately, often will not manage to answer it on the same day. (Sometimes, he will be out of town and it will take him even longer.)
It is very important to us that you voice your concerns about any aspect of the class as soon as they arise. Please send an e-mail to the instructor or talk to us in person. We will accept anonymous notes (either on paper or via email from any free "on-the-fly" email account) and treat them seriously, as long as they are sincere and constructive. Your comments will have an effect on the class, so do not hesitate to provide them.
Artificial Intelligence is a fun topic, and we hope that all of us will have lots of fun!
Sven (yes, please feel free to call the instructor by his first name, which is of Scandinavian origin and means "young") and the TA