Print

2023　Advanced Methodology of Mathematical and Computational Analysis I

Font size  SML

Register update notification mail Add to favorite lecture list
Academic unit or major
Graduate major in Technology and Innovation Management
Instructor(s)
Ikeda Shintaro 
Class Format
Lecture / Exercise    (Livestream)
Media-enhanced courses
Day/Period(Room No.)
Mon11-12()  
Group
-
Course number
TIM.A538
Credits
1
Academic year
2023
Offered quarter
3Q
Syllabus updated
2023/4/10
Lecture notes updated
-
Language used
Japanese
Access Index

Course description and aims

　Dramatic advances in deep learning technology have brought about major changes not only in the development of academic research, but also in industry and society. Image recognition AI and game AI have already been shown to surpass human capabilities, and new algorithms are constantly being proposed and put to practical use in classical optimization and search technologies. New algorithms are being proposed and put to practical use in classical optimization and search techniques. In addition, quantum computing is gradually being applied in limited situations (e.g., combinatorial optimization). In the future, basic knowledge of these advanced technologies and understanding of their technical limitations will be required not only for engineers in the field but also for management persons
　Since this lecture is targeted at students studying technology management, we will avoid detailed theoretical explanations of mathematics as much as possible, and aim to acquire a wide range of basic knowledge related to deep learning. In addition, by creating simple models through programming practice, the objective is for the participants to correctly understand the merits and demerits of deep learning and the scope of its application.
The programming practice in this lecture is intended to be at a level corresponding to beginners of Python. This lecture is designed to be easy to learn even for beginners of machine learning, and the students will acquire the principles of algorithms and the ability to implement simple methods based on the application examples of each method.

Student learning outcomes

By taking this lecture, students will understand and acquire the following.
(1) Understand the development trend, various problems, and business cases of deep learning.
(2) To understand the outline of machine learning algorithms for image recognition and natural language processing.
(3) To correctly understand the advantages and disadvantages of deep learning, its application scope, and technical limitations.
(4) Acquire the ability to implement deep learning using Python.

Course taught by instructors with work experience

✔ Applicable How instructors' work experience benefits the course
The teacher in charge of this lecture has been conducting research on artificial intelligence and mathematical optimization for 10 years, and at the same time, has been developing software for machine learning at a company he founded.

Keywords

Deep learning, image recognition, natural language processing, Python, programming

Competencies that will be developed

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

Class flow

The class will consist mainly of lectures, with some Python programming practice and group work, using the browser-based development environment (Google Colab).

Course schedule/Required learning

  Course schedule Required learning
Class 1 Guidance, trends in the development of artificial intelligence and various issues in the field of artificial intelligence To understand the purpose of this lecture and the technological transition of artificial intelligence. Students will also understand various problems in the field of artificial intelligence, such as frame problems and symbol grounding problems.
Class 2 Classification and evaluation metrics of machine learning methods To understand the characteristics and classification of various artificial intelligence algorithms and the evaluation index of prediction accuracy.
Class 3 Neural Networks and Deep Learning To understand the basic principles and deepening of neural networks.
Class 4 Time Series Data Processing (Programming) To understand the principles and applications of recurrent neural networks; to implement a weather prediction AI using Python programming.
Class 5 Computer Vision and Natural Language Processing To understand the principles and applications of convolutional neural networks. To understand natural language processing such as neural machine translation and speech synthesis.
Class 6 Bayesian statistics and machine learning, group work 1 Deepen your understanding of Bayesians as implemented in spam filters, etc.
Class 7 Group presentation Present business ideas related to artificial intelligence.

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

To enhance effective learning, students are encouraged to spend approximately 100 minutes preparing for class and another 100 minutes reviewing class content afterwards (including assignments) for each class.
They should do so by referring to textbooks and other course material.

Textbook(s)

Nothing

Reference books, course materials, etc.

Related materials will be distributed during the lecture.
In addition, the following books may be referred to as necessary.
Uji Igari et al., Deep Learning Textbook: G-test (Generalist) Official Textbook, 2nd Edition, Shoei-sha, April 27, 2021.

Assessment criteria and methods

Evaluation will be based on participation in lectures and programming practice (50%) and submission of reports (50%).

Related courses

  • TIM.A405 ： Methodology of Mathematical and Computational Analysis I
  • TIM.A406 ： Methodology of Mathematical and Computational Analysis II
  • TIM.A539 ： Advanced Methodology of Mathematical and Computational Analysi II

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

Nothing

Other

Please check T2SCHOLA for class materials.

Page Top