An artificial intelligence system consists of many elemental techniques. In this course, students first learn how to formulate simple problems and how to search for solutions to them. Then, they learn how to explicitly represent knowledge and inference procedure based on the representation. Next, they learn about planning for efficient inference. Finally they learn machine learning in which computers automatically acquire knowledge.
The goals of this course are to understand the necessity of artificial intelligence systems which assist human intellectual activities in information society and to learn the elemental techniques to build them. In particular, its aim is to learn how to represent the process of human intellectual activities and to infer based on the representation. In this course, students learn the methodology of such areas as search, knowledge representation, inference, planning, machine learning, which are necessary for making artificial intelligence systems, and make it as the basis to apply the systems to the real world.
search, semantic network, frame representation, production system, resolution principle, inference, planning, linear classification, neural network, decision tree,
✔ Specialist skills | Intercultural skills | Communication skills | Critical thinking skills | ✔ Practical and/or problem-solving skills |
1) At the beginning of each class, the previous class is reviewed, and solutions to exercise problems are given.
2) Exercise problems of the day are given towards the end of class.
Course schedule | Required learning | |
---|---|---|
Class 1 | Introduction of artificial intelligence | Specified in the class. |
Class 2 | Search 1: Problem formulation and graph search | Specified in the class. |
Class 3 | Search 2:Heuristic search and A* search | Specified in the class. |
Class 4 | Search 3: Game (min-max, alpha-beta, reversi, chess) | Specified in the class. |
Class 5 | Knowledge representation 1: Semantic network and frame representation | Specified in the class. |
Class 6 | Knowledge representation 2: Production system | Specified in the class. |
Class 7 | Inference 1: Inference using resolution principle | Specified in the class. |
Class 8 | Inference 2: Forward and backward inference, default inference | Specified in the class. |
Class 9 | Inference 3: Probabilistic inference (Bayesian network) | Specified in the class. |
Class 10 | Plannning 1: GPS, Hierarchical planning | Specified in the class. |
Class 11 | Plannning 2: Partial-order planning, reactive planning | Specified in the class. |
Class 12 | Machine Learning: Linear classifier | Specified in the class. |
Class 13 | Machine Learning 2: Neural network | Specified in the class. |
Class 14 | Machine learning 3: Decision trees, misc. | Specified in the class. |
Class 15 | Applications and future outlook | Specified in the class. |
Not specified
S. Russel and P. Norvig, "Artificial Intelligence: A Modern Approach", Prentice Hall.
Students' course scores are based on exercise problems (20%), and final exam (80%).
No prerequisites.