Print

2023　Machine Learning (ICT)

Font size  SML

Register update notification mail Add to favorite lecture list
Academic unit or major
Undergraduate major in Information and Communications Engineering
Instructor(s)
Suzuki Kenji 
Class Format
Lecture    (Face-to-face)
Media-enhanced courses
Day/Period(Room No.)
Tue3-4(M-B07(H101))  Fri3-4(M-B07(H101))  
Group
-
Course number
ICT.S311
Credits
2
Academic year
2023
Offered quarter
3Q
Syllabus updated
2023/9/29
Lecture notes updated
-
Language used
Japanese
Access Index

Course description and aims

In this course, you will learn the mathematics, basic principles, methods of machine learning and their applications. To make those machine learning methods your own knowledge, you will practice programming of them and how to implement them by using machine learning libraries. First, you will learn the fundamental knowledge on and basic principles of machine learning including supervised & unsupervised learning, classification & regression, and ill-posed problem & optimization. Next, you will learn major machine learning models and methods and the basic principles, features, and implementations of them. Those models and methods include linear discriminant analysis, principal component analysis, training and validation methods, clustering, support vector machines, neural networks, and deep learning models.

Student learning outcomes

To learn the fundamentals and basic principles of machine learning and make them your own knowledge and skills to solve machine learning problems. To learn major machine learning methods and make them your own knowledge to apply them properly. To learn how to implement machine learning methods by programming.

Keywords

Least square method, Linear discriminant analysis (LDA), Principal component analysis (PCA), K-means method, Bayes estimation, Neural networks, Backpropagation, Support vector machine (SVM), Ensemble learning, Deep learning, Convolutional neural network (CNN).

Competencies that will be developed

✔ Specialist skills Intercultural skills Communication skills Critical thinking skills ✔ Practical and/or problem-solving skills

Class flow

In the first three classes, you will learn the fundamental mathematics and basic principles of machine learning. In the following eight classes, you will learn major machine learning models and methods. In each class, you will practice the implementations of the methods by programming them with machine learning libraries. In the last three classes, you will learn and practice deep learning.

Course schedule/Required learning

  Course schedule Required learning
Class 1 Introduction to the course To learn the summary of the course, and the basic knowledge on, the field of, and applications of machine learning.
Class 2 Principle of learning 1 To understand supervised & unsupervised learning, classification & regression, ill-posed problem & optimization, and the least square method.
Class 3 Principle of learning 2 To understand the degrees of freedom in a model & different types of errors, over-fitting & regularization, and the cause of dimensionality.
Class 4 Linear discriminant analysis (LDA) To learn the linear discriminant, within-class variance & between-class variance, and Fisher’s linear discriminant analysis.
Class 5 Principal component analysis (PCA) To learn the mathematics behind PCA, PCA for dimensionality reduction, data compression with PCA, and differences from LDA.
Class 6 Methodology of training, validation, and testing To study the methodology of training, validation, and testing, and performance metrics in machine learning.
Class 7 Unsupervised learning and Bayesian methods To learn clustering algorithms including the K-means algorithm, maximum likelihood estimation, and expectation-maximization (EM) algorithm.
Class 8 Neural networks 1 To learn biological neural networks and their models, the history of neural networks, a neuron model, and the perceptron.
Class 9 Neural networks 2 To learn the multilayer perceptron, activation functions, and the error back propagation and its derivation.
Class 10 Neural networks 3 To learn how to interpret neural networks, how to design neural networks, and neural network applications.
Class 11 Support vector machines (SVMs), pre-processing of data, and fusing multiple models To learn linear and nonlinear SVMs, the kernel tricks, a missing data issue, normalization of data, and ensemble learning.
Class 12 Overview of deep learning To learn convolutional neural networks (CNNs), LeNet, AlexNet, VGG-16, ResNet, and MTANN.
Class 13 Computation of deep learning To learn the convolution, pooling, and softmax layer in convolutional neural networks (CNNs).
Class 14 Applications of deep learning To learn deep learning applications to image processing, object detection, segmentation, super-resolution, voice recognition, natural language processing, medical image processing, and medical image diagnosis.

Out-of-Class Study Time (Preparation and Review)

To enhance effective learning, students are encouraged to spend approximately 100 minutes for preparing for a class and another 100 minutes for reviewing the class contents (and assignments) after each class by using the textbook, reference books, and other course materials.

Textbook(s)

A textbook to study deep learning lectures and practices in the last three classes in this course.
J. Krohn (author), K. Suzuki (supervisor of a translation), M. Shimizu (translator),「Python, TensorFlowで実践する深層学習入門: しくみの理解と応用」* , Tokyo Kagaku Dojin Publishing, Ltd., 278 pp., ISBN：9784807920389, 2022.
*This book is the Japanese edition of “Deep Learning Illustrated”.

Reference books, course materials, etc.

A reference book to some machine learning methods in this course.
M. Sugiyama, 「イラストで学ぶ　機械学習」最小二乗法による識別モデル学習を中心に, Kodansha ltd., ISBN：978-4-06-153821-4, 2013

Assessment criteria and methods

Final examination and reports for assignments and programming exercises.

Related courses

  • LAS.M102 ： Linear Algebra I / Recitation
  • LAS.M101 ： Calculus I / Recitation
  • LAS.M105 ： Calculus II

Prerequisites (i.e., required knowledge, skills, courses, etc.)

Basic knowledge on differential and integral calculus.

Page Top