| Course title | |||||
| 人工知能 [Artificial Intelligence] | |||||
| Course category | technology speciality courses | Requirement | Credit | 2 | |
| Department | Year | 3～4 | Semester | 1st | |
| Course type | 1st | Course code | 023664 | ||
| Instructor(s) | |||||
| 藤田 桂英 [FUJITA Katsuhide] | |||||
| Facility affiliation | Faculty of Engineering | Office | afjgxte/L1151 | Email address | |
| Course description |
|
This lecture demonstrates some intelligence algorithms and AI systems. These topics range from Base technology to state-of-art one. In the first half of the class, the area of AI which was studied actively in 1980s’ is shown. For example, search algorithms are the one of an important problem-solving technique, and can solve the particular problems based on the heuristics. Also, the knowledge representations and Logic can be learned from leading to Prolog. In the second half of the class, the topics related to Natural Language Processing, Web Intelligence, and Software Agents are learned as explaining the algorithms and idea related to them. |
| Expected Learning |
|
By the end of this course, we expect that the whole knowledge related to AI and the ability to generate the AI system can be learned. See the Curriculum maps. |
| Course schedule |
|
(1st half of class: 1－7) - What’s AI?: History of AI, Turing Test, Learning Machine, AI and related area - Search and Problem Solving: State Space, Depth-first Search, Width-first Search, Recursion Algorithm, Best-first Search - AI with Game: Variety of Games, Game Tree, AND/OR Tree, min-max strategy, alpha-beta cutoff - Bio-Inspired Algorithm: Genetic Algorithm, Hill-Climbing, Simulated Annealing - Knowledge Representation: Semantic Network, Frame, Hierarchy, Inheritance, Relationship between Knowledge Representation and NLP - Predicate Logic: Propositional Logic, Predicate Logic, Resolution, Logical Reasoning （2nd half of class: 8-14） - Web Intelligence: Search Engine (Indexing, Ranking, Crawling), Community Formation (Complex Network), Recommendation System and Collective Intelligence, Web programing - Reinforcement Learning: Markov decision process, Q-Learning, Deep Q-Network - Software Agent: What’s Agent? Cooperation (Problem Solving, Search, Action Selection), Negotiation, Market-based Mechanism, Auction, Complex System |
| Prerequisites |
|
It is better to take "Introduction to Algorithms," "Algorithms : Laboratory Exercises," and "Algorithms" previously. You should prepare and review the contents of this lecture in addition to the standard hours (30 hours) of lectures, using teaching materials and reference books. |
| Required Text(s) and Materials |
| None |
| References |
| - S.J.Russell, P.Norvig, “Artificial Intelligence: A Modern Approach” |
| Assessment/Grading |
|
The grade evaluation in this online class is premised on all attendances, and comprehensively evaluates the attitude to learn, quizzes, report, and online tests. The standard study time set by the our university is required to get the grade. The rate of evaluation is as follows: Small Quiz, reports, and Examination (100%) Grade will be given according to the following criteria by comprehensive evaluation: S: 90 points or more, A: 80 or more and less than 90 points, B: 70 or more and less than 80 points, C: 60 or more and less than 70 points. |
| Message from instructor(s) |
| Through the AI class, you can learn the way that computers solve the complex problems with the intelligence. In addition, you can learn some important AI algorithms which work in the society. |
| Course keywords |
| Problem Solving, Game-Tree Search, Knowledge Representation, Natural Language Processing, Web Intelligence, Software Agents |
| Office hours |
| Please e-mail me when you have questions. |
| Remarks 1 |
| Remarks 2 |
| Related URL |
| Lecture Language |
| Japanese |
| Language Subject |
| Last update |
| 2/2/2021 11:01:50 AM |