TU Delft
print this page print this page     
2015/2016 Electrical Engineering, Mathematics and Computer Science Bachelor Computer Science and Engineering
Logic Based AI
Responsible Instructor
Name E-mail
Dr. N. Bulling    N.Bulling@tudelft.nl
Dr. K.V. Hindriks    K.V.Hindriks@tudelft.nl
Contact Hours / Week x/x/x/x
0/0/4/0 hc; 0/0/2/0 prac
Education Period
Start Education
Exam Period
Course Language
Course Contents
AI (Artificial Intelligence) techniques that are discussed in this course are knowledge representation and reasoning techniques, and multi-agent technology. Students are taught how to develop a multi-agent system that uses knowledge representation to reason about the environment in which the multi-agent system operates. In the Project Multi-agent systems following this course, students develop, based on the knowledge gained in this course, a team of intelligent agents that drive bots in the game Unreal Tournament.
Study Goals
1. Agents and Multi-Agent Systems
Agent, multi-agent system, mental state, beliefs, goals, actions, communication, coordination. The student is able to define the elementary concepts, and to apply their relevant aspects in the design of programs.
2. Agent-Environment Interaction
Action, percept. The student is capable of analysing the environment in which the agent operates, and to identify the actions and relevant percepts that are available. The student is able to use the classification, actions and percepts as the basis for the design of an agent.
3. Agent and Multi-Agent Program
Action rules, modules, MAS file, launch rules. The student is able to describe the relevant programming constructs. In addition, the student is able to apply the constructs to write a multi-agent and agent programs.
4. Basic Concepts Prolog
Facts, rules, clause, queries, rule-based reasoning. The student is able to define the basic concepts in Prolog and describe the relationship between these concepts. Based on these concepts the student is able to solve simple problems in Prolog.
5. Prolog Programming
Negation as failure, cut, recursion, lists. The student is able to apply constructs and these techniques to write a Prolog program. The student is able to solve problems by using a combination of negation as failure, recursion and the use of lists. In particular, the student is able to implement some search algorithms in Prolog.
6. Reasoning in Logic Programming
Unification, backtracking, depth-first search, linear search, backward chaining. The student is able to explain the computational model of logic and reasoning in Prolog and to use these concepts. Simple tasks with unification of terms can be made by the student. The student is able to construct a derivation of a unification (resolution).
7. Develop a MAS
The student is able to build a relatively simple multi-agent system. Concepts relating to systems of rational agents are introduced to make complex decisions.
8. Rule-based system
The student is able to explain what a rule-based system (Expert System) is.
Education Method
Lectures, lab sessions (with computer/ laptops). Attendance at labs is obligatory.
Computer Use
Computer (laptops) are used during labs.
Literature and Study Materials
Slides and Script.

Intelligent Systems for Engineers and Scientists, third edition (Adrian A. Hopgood)

Learn Prolog Now! (Blackburn et al.),
GOAL programming guide

The material is available online.
During contact exercises (self-assessment): During the lectures, new concepts are practiced on the basis of simple tasks/ examples.

During contact exercises (self-assessment): During practical classes with computers (laptops) student assistants help students in solving problems. With each task an oral test is conducted to test whether the student has sufficient understanding of the submitted solution.

The course is assessed through an exam and a practical part.
Permitted Materials during Tests
For the exam no material is allowed (pen and paper exam).
All students are obligated to enrol in the course in BlackBoard. All announcements will be made via Blackboard. All material will be provided and needs to be submitted via Blackboard. Also, groups for the practical part will be announced via Blackboard.

The course Logic-Based Artificial Intelligence Project and Multi-Agent Systems are integrated to a large extent. At the lectures the theory and techniques are discussed which are used in the project.
The final grade is a weighted average of the exam grade (75%) and the grade of the practical part (25%).

Practical part: Students get 4x a practical assignment that must be made in pairs. The assignments are graded with a mark. The grade for the practical part is the average of the marks for the assignments. There must be a minimum of 5 on average obtained for the assignments to pass the practical part.

The exam consists of open questions about all the material of the course. To pass the exam the grade must be at least a 5.

The exam can be retaken once per year. There is no second chance for the practical assignments.