This course teaches advanced technologies of artificial intelligence. This course consists of two parts. The topics of the first part include evolutionary computation and reinforcement learning. In the second part, students will learn Knowledge Representation and Reasoning, which is dedicated to representing information about the world in a form that a computer system can utilize to solve complex tasks.
The aims of this course is to enable students 1) to acquire knowledge on evolutionary computation, reinforcement learning and symbolic knowledge representation, and 2) to apply the knowledge to solve real-world problems.
By the end of this course, students will learn the following:
1) Evolutionary computation techniques and how to apply them to real-world problems.
2) Reinforcement learning techniques and how to apply them to real-world problems.
3) Knowledge representation techniques, in particular, logic-based representation.
4) Advanced Reasoning techniques, including abduction and induction.
evolutionary computation, reinforcement learning, black-box optimization, multiobjective optimization, weak-supervised learning, knowledge representation, deductive reasoning, inductive reasoning, abductive reasoning, constraint satisfaction, and planning.
|Intercultural skills||Communication skills||Specialist skills||Critical thinking skills||Practical and/or problem-solving skills|
Every class consists of a lecture using the slides and the exercise. Students are required to download the materials of lecture and read them before the class.
|Course schedule||Required learning|
|Class 1||Introduction||Understand the background and the aim of the course.|
|Class 2||Evolutionary computation for function optimization: Genetic algorithms||Understand function optimization and genetic algorithms.|
|Class 3||Evolutionary computation for function optimization : Evolution strategies||Understand evolution strategies.|
|Class 4||Evolutionary computation for combinatorial optimization||Understand genetic algorithms for combinatorial optimization.|
|Class 5||Evolutionary computation for multiobjective optimization||Understand multiobjective optimization and evolutionary multiobjective optimization.|
|Class 6||Reinforcement learning based on value functions||Understand Markov decision processes and reinforcement learning based on value functions.|
|Class 7||Reinforcement learning based on direct policy search||Understand partially observable Markov decision processes and direct policy search methods.|
|Class 8||Deep reinforcement learning||Understand deep reinforcement learning techniques.|
|Class 9||Knowledge representation and reasoning in AI||Understand the role of knowledge representation in symbolic and neural levels.|
|Class 10||Logic-based knowledge representation||Understand the syntax, semantics, and logical consequences.|
|Class 11||Propositional logic and constraint programming||Understand the satisfiability checking (SAT) and constraint satisfaction problems (CSP).|
|Class 12||Predicate logic and logic programming||Understand the resolution principle, fixpoint semantics, and procedural semantics.|
|Class 13||Commonsense reasoning||Understand how to represent and reason commonsense knowledge, nonmonotonic reasoning, and answer set programming.|
|Class 14||Reasoning about action and change||Understand the situation calculus, event calculus. planning, and temporal logic.|
|Class 15||Abduction and induction||Understand hypothesis finding, program synthesis, and inductive logic programming.|
No textbook is set. Materials are distributed before each lesson.
Artificial Intelligence - A Modern Approach (Third Edition, Prentice Hall), and so on.
Students’ course scores are based on the first part (50%) and the second one (50%). In the first part, students’ scores are based on exercise problems/homework in classes 2 to 7 (20%) and exercise problems for assessing students’ level of understanding in class 8 (80%). In the second part, homework, intermediate presentations and final reports are put together and are then evaluated.