• If you are Japanese student, please also register at the NU-EMI project for this course. 
  • We will use Discord for communication and for sharing materials & code. 
  • We will use TACT for homework submission. Please contact me if you are not in TACT.
​
この講座は機械学習に用いられる数学の入門です。授業は英語で行われます。この科目に興味のある人や英語で行われる授業に参加したい人は、名古屋大学の学生であれば誰でも大歓迎です。この講座の受講する場合、線形代数と微積分の基礎的な内容に慣れている方が望ましいですが、最初の講義でPythonを基礎から学びますのでプログラミングの知識は不要です。この講座の履修登録はNU EMI​で済ませてください。英語に不安のある学生には、日本人のTAによるサポートもあります。

​This course is targeted at any student of Nagoya University who is interested in some of the mathematics used in machine learning. As a background I expect that the students have some background in Linear Algebra and Calculus (e.g. Linear Algebra I & Calculus I from the G30 Program). I do not expect that the students can already program, but if you do not have any Python experience you might need to do a basic course by yourself at the beginning (or in the summer break). If you took the Data Science Exercise B then this course is perfect for you. If you have any questions or suggestions on the course please feel free to contact me. 

Materials
Content (tentative)
Machine learning became a popular and really broad field in recent years. Machine learning algorithms are used in a wide variety of applications, such as email filtering, computer vision, medicine, language translation, computer games, economic, etc.. The goal of this course is to give a brief introduction into machine learning with a focus on the mathematical tools used. 
We will probably cover at least the following topics:
  • Introduction to machine learning and programming in Python
  • Supervised learning: Linear & Logistic regression,  Neural networks
  • Language models: Transformers
  • Unsupervised learning: PCA, Autoencoder
  • Reinforcement Learning: Q-Learning
    ​
Grading
The grade will we based on homework (70%) submissions and one final project (30%). 
The credits follow the new rule for SML classes. (SML=Special mathematics lecture)
Please contact me via email if you have any questions.

Final Project
Choose a Nagoya or Japan-related topic to solve using machine learning. Develop a model to address a specific problem or provide insights into life, business, environment, culture, etc.
Group Size: 1-3 members
Submission: Preferably in a Google Colab notebook (other options possible with prior approval). 
Presentation (January): 5-10 slides covering:
  • Problem Statement
  • Data Collection
  • Data Exploration & Visualization
  • Model Building & Evaluation
  • Conclusion​

Example Projects from Fall 2023
Here are some examples of last year's semester projects (in random order)
Course schedule (tentative) 
Lecture:
  • Time: Monday 6th period (18:15-19:45). First lecture: October 7th.
  • Room: A250 (Computer room) in Science Building A (​D3 (2) on the Nagoya University Campus map)
Tutorial: 
  • Time: Tuesday 6th period (18:15-19:45)
  • Room: A250 (Same as the lecture)

Course Overview:
The following gives an overview of the topics we will cover each week. 
Week 01 (10/07-10/13): Introduction
Week 02 (10/14-10/20): Linear Regression
Week 03 (10/21-10/27): Linear Regression II & Logistic Regression
Week 04 (10/28-11/03):  Logistic Regression II & Neural Networks I
Week 05 (11/04-11/10): No lecture on 4th November (Culture day) 
Week 06 (11/11-11/17): Neural Networks II
Week 07 (11/18-11/24): Neural Networks III
Week 08 (11/25-12/01): Neural Networks IV
Week 09 (12/02-12/08): Q-Learning I
Week 10 (12/09-12/15): Q-Learning II & k-means clustering
Week 11 (12/16-12/22): Deep Q-Learning
Week 12 (12/23-12/29): Christmath Quiz & Language Models I
☃️ Winter Vacation (12/28-01/08) ☃️ 
Week 13 (01/06-01/12): Language Models II (Thursday 9th January)
Week 14 (01/13-01/19): No lecture on 13th January (Coming-of-Age Day)
Week 15 (01/20-01/26): Language Models III & Presentations
Week 16 (01/27-02/06): -

Last update: 9th January 2025.