Skip navigation
The Australian National University
Artificial Intelligence COMP3620/COMP6320

Artificial Intelligence COMP3620/COMP6320

Welcome to the AI course at the ANU !

The course covers a range of different topics within the field of AI, each taught by a specialist in that area. The Introduction and Search is taught by Sylvie Thiebaux who is also the convenor and general contact person of the course, Knowledge Representation and Reasoning by Miquel Ramirez, Planning by Sylvie again and Reinforcement Learning by Marcus Hutter. The main website for communication during the course is the wattle page.


Offered By: The AI Group @ Research School of Computer Science @ Australian National University
Offered In: First Semester, 2015
Lecturers: Prof. Sylvie Thiebaux (Convenor), Dr. Miquel Ramirez, Prof. Marcus Hutter
Tutors/Labs: Dr. Miquel Ramirez, Dr. Enrico Scala
Target: Undergraduate (COMP3620) and Graduate (COMP6320) students. Others welcome.
Enrollment: Undergraduates &Masters: The usual way via ISIS. Honors&Others: Contact lecturer.
Admin: Ms. Bindi Mamouney.
Unit Value: 6 units
Time Table: See
Assessment: Assignments (50%); Exam (50%). Students must achieve at least 50% overall and 40% on the final exam to pass.
Textbook: Course textbook is "Artificial Intelligence - A Modern Approach"(3rd edition), by Stuart Russell and Peter Norvig (Prentice Hall). This book gives a comprehensive tour of AI, and not all of it is part of the course material. Details about which chapters are included and additional recommended reading will be given by the lecturers for each topic.
Programs and Courses page:
Wattle page:
This page:

Prerequisites: What you really need is some background in computer science and programming, as well as some knowledge of logic. As stated in programs and courses, this has been translated into the following formal requirements:
  • computer science/programming: COMP1100 or COMP1130 and COMP1110 or COMP1140 or COMP1510
  • logic: COMP2620 or COMP2600
Contact the convenor (Sylvie Thiebaux) if you do not meet the formal requirements but is still interested in taking the course.

Learning Outcomes

After completing this course, students should be able to:
  • Identify problems that are amenable to solution by AI methods, and which AI methods may be suited to solving a given problem.
  • Formalise a given problem in the language/framework of different AI methods (e.g., as a search problem, as a constraint satisfaction problem, as a planning problem, as a Markov decision process, etc).
  • Implement basic AI algorithms (e.g., standard search algorithms or dynamic programming).
  • Design and carry out an empirical evaluation of different algorithms on a problem formalisation, and state the conclusions that the evaluation supports.


There will be five assignments, one on search, on knowledge representation and reasoning, one on planning, one on reinforcement learning, and in addition, one preliminary assignment in which you will familiarise yourself with the programming language we will use: Python. These make up half your final mark. The other half is the exam. The assignments are both released and submitted through Wattle.


Assessment consists of five assignments and a final exam. The final mark is determined 50% by assignments, 50% by the exam. All assignments have equal weight. Late assignments are not accepted. However, each student will have a total extension budget of 72h (3 days) for the whole assignment series. This budget can be spent without justification nor prior request. Once the budget is exhausted, 100% penalty will apply. The minimum final mark required for a pass grade is 50% as usual. Note, however, that in order to obtain a passing grade, students must reach a minimum of 40% marks on the exam.

Updated:  03 June 2015 / Responsible Officer:   JavaScript must be enabled to display this email address. / Page Contact:   JavaScript must be enabled to display this email address. / Powered by: Snorkel 1.4