Course title
L04403003
Artificial Intelligence

 igarashi harukazu Click to show questionnaire result at 2019
Course description
In this course, we take an engineered approach where the goal of “Artificial Intelligence” (AI) is reproduction of human intelligence by computer. Recently, the area of AI is extending from engineering such as information processing and robotics to logic, mathematics, biology, physiology, cognitive science, physics and social science, and is growing even now. Therefore, you need a wide range of knowledge, ability to think logically and information skills to understand the mathematical foundation and apply AI algorithms to practical systems in the real world. This course presents general basic theory and algorithms that can be applied to practical AI problems.
Purpose of class
人工知能の基本的な概念や理論を学ぶと共に、実装上のアルゴリズムや具体的な例題への適用例を理解する。
Goals and objectives
  1. To understand algorithms for tree and graph search profoundly enough to explain the algorithms precisely and solve simple AI problems.
  2. To understand approximation algorithms for combinatorial optimization problems and explain their basic principles.
  3. To understand methods of knowledge representation and inference that includes predicate logic, production systems, frame theory and Bayesian networks profoundly enough to explain the algorithms precisely and solve simple inference problems. These methods are used for constructing expert systems.
  4. To understand some basic algorithms of machine learning and explain their learning rules. The basic algorithms include ID3, neural network models and Q-learning.
Language
Japanese
Class schedule

 Class schedule HW assignments (Including preparation and review of the class.) Amount of Time Required
1. What is AI : Definition, research fields and history of Artificial Intelligence Read this syllabus. 100minutes
2. Blind search and heuristic search : Tree search, state space, depth-first search, breadth-first search, optimal search, best-first search and A*algorithm Review depth-first search and breadth-first search presented in “Data Structure and Algorithms 2” (L0694500). 200minutes
3. AND/OR graph search : Evaluation phase and Expansion phase Review the content of the 2nd lecture. 100minutes
4. Game tree search : minimax algorithm and αβ search Check the reference (1) or the Internet to understand “minimax algorithm”. 100minutes
5. Approximate solution of combinatorial optimization problems (I) : Simulated annealing Use the Internet to find out what a traveling salesman problem is. 100minutes
6. Approximate solution of combinatorial optimization problems (II) : Genetic algorithm Use the Internet and references to understand Genetic Algorithms. 200minutes
7. Logic (I) : First-order predicate calculus Understand definitions of propositional logic and predicate logic by reading the reference (1). 100minutes
8. Logic (II) : Clause form, syllogisms and resolution principle Review meanings of symbols and technical terms presented at the 7th lecture. 100minutes
9. Production system : IF-THEN rule and blackboard model

Frame theory : frame representation (is-a, part-of), class and daemon 		Understand objectives of expert systems by reading references (2) and (3). 100minutes
Understand objectives of frame theory by reading references (2) and (3). 100minutes
10. Reasoning with uncertainty : Uncertainty factor and Bayesian network Review a basic knowledge of probability. 200minutes
11. Inductive learning (I) : Learning of decision tree (ID3 algorithm) Review a basic knowledge of information entropy. 200minutes
12. Inductive learning (II) : Multilayer neural network model (Error backpropagation algorithm) Review a basic knowledge of partial differentiation. 200minutes
13. Reinforcement learning : Markov decision processes and Q-learning algorithm Use the Internet and references to understand reinforcement learning. 200minutes
14. Final exam, Q&A Review the contents of the 1st to 14th lectures. Train yourself to solve concrete problems if the general formula is presented. 650minutes
Total. - - 2650minutes
Relationship between 'Goals and Objectives' and 'Course Outcomes'

 Final exam Total.
1. 25% 25%
2. 25% 25%
3. 25% 25%
4. 25% 25%
Total. 100% -
Evaluation method and criteria
Final exam (100%). Over 60% is acceptable.
Textbooks and reference materials
Required Textbook: Not required.

Reference books:
(1) M. Shirai, “AI theory：enlarged edition,” Coronasha (in Japanese).
(2) N. Babaguchi and S. Yamada, “Basis of AI:2nd edition,” Shokodo (in Japanese).
(3) K. Nitta,”Introduction to AI,” Baifukan (in Japanese)
Prerequisites
Recommendation : “Data Structure and Algorithms 2”(L0694500).
Office hours and How to contact professors for questions
  • Anytime except lecture hours at my office (Room no.14M32 or 14K30). E-mail contact is also available to ask questions.
Regionally-oriented
Non-regionally-oriented course
Development of social and professional independence
  • Course that cultivates an ability for utilizing knowledge
Active-learning course
N/A
Course by professor with work experience
Work experience Work experience and relevance to the course content if applicatable
N/A N/A
Education related SDGs:the Sustainable Development Goals
  • 9.INDUSTRY, INNOVATION AND INFRASTRUCTURE
  • 12.RESPONSIBLE CONSUMPTION & PRODUCTION
Last modified : Sat Mar 21 12:29:46 JST 2020