IISTIIST (Institute of Integrated Science and Technology)
HUI500D1（人間情報学 / Human informatics 500）IIST Special Lecture 4IIST Special Lecture 4
Machine Learning
余 恪平Keping YU
授業コードなどClass code etc
| 学部・研究科Faculty/Graduate school | IISTIIST (Institute of Integrated Science and Technology) |
| 添付ファイル名Attached documents | |
| 年度Year | 2024 |
| 授業コードClass code | YD968 |
| 旧授業コードPrevious Class code | |
| 旧科目名Previous Class title | |
| 開講時期Term | 春学期授業/Spring |
| 曜日・時限Day/Period | 水3/Wed.3 |
| 科目種別Class Type | |
| キャンパスCampus | 小金井 / Koganei |
| 教室名称Classroom name | 小西館‐ゼミ室1／W Seminar 1(6F) |
| 配当年次Grade | |
| 単位数Credit(s) | 2 |
| 備考（履修条件等）Notes | |
| 実務経験のある教員による授業科目Class taught by instructors with practical experience |
すべて開くShow all
すべて閉じるHide All
授業の概要と目的（何を学ぶか）Outline and objectives
This course provides a broad introduction to machine learning. Topics include regression, classification, meta learning, reinforcement learning, network compression, and so on.
到達目標Goal
The students will get to understand the key techniques in machine learning and gain practice implementing them and getting them to work for themselves.
授業で使用する言語Default language used in class
英語 / English
授業の進め方と方法Method(s)（学期の途中で変更になる場合には、別途提示します。 /If the Method(s) is changed, we will announce the details of any changes. ）
It will be given in the Learning Management System. Please confirm the announcement from the Learning Management System.
The students will be asked to submit projects on the basic machine learning problems and consider their own solutions. The students will be encouraged not only to learn the knowledge but also to think about how it can be used to solve real problems.
アクティブラーニング（グループディスカッション、ディベート等）の実施Active learning in class (Group discussion, Debate.etc.)
あり / Yes
フィールドワーク（学外での実習等）の実施Fieldwork in class
あり / Yes
授業計画Schedule
授業形態/methods of teaching：対面/face to face
※各回の授業形態は予定です。教員の指示に従ってください。
1[対面/face to face]：Introduction
What is machine learning? What is supervised learning? What is unsupervised learning?
2[対面/face to face]：Linear regression with one variable
Model representation, cost function, gradient descent for linear regression.
3[対面/face to face]：Linear regression with multiple variables
Multiple features, gradient descent for multiple variables.
4[対面/face to face]：Logistic regression
Classification, hypothesis representation, cost function,
5[対面/face to face]：Regularization
The problem of overfitting, regularized linear regression.
6[対面/face to face]：Neural networks
Non-linear hypotheses, neurons and the brain, model representation.
7[対面/face to face]：Back-propagation algorithm.
Back-propagation algorithm, gradient checking, random initialization.
8[対面/face to face]：Machine learning system design
Examples of implementing machine learning system.
9[対面/face to face]：Meta learning
Introduction and case study of meta learning
10[対面/face to face]：Reinforcement learning
Introduction of reinforcement learning
11[対面/face to face]：Anomaly detection
Introduction and case study of anomaly detection
12[対面/face to face]：Network compression
Introduction and examples of network compression
13[対面/face to face]：More examples and summary
Introduce more examples and summarize the lecture
14[対面/face to face]：Project presentation
Students report their projects.
授業時間外の学習（準備学習・復習・宿題等）Work to be done outside of class (preparation, etc.)
【Preparatory study and review time for this class are 4 hours each.】Homework and final presentation.
テキスト（教科書）Textbooks
Handouts and prints will be distributed.
参考書References
Ian Goodfellow, "Deep learning".
成績評価の方法と基準Grading criteria
Homeworks 30% + Final report 70%
学生の意見等からの気づきChanges following student comments
None in particular.