DATE2024.01.10 #Events
Disclaimer: partially machine translated using DeepL and may contain errors.
The 36th Public Lecture, School of Science, The University of Tokyo
The World Opened Up by Science
At School of Science, The University of Tokyo, researchers who are fascinated by the unsolved mysteries of the natural world are working day and night in pursuit of the answers. The discoveries and science and technology produced from these efforts not only make academic contributions, but also lead to the cornerstones of cutting-edge fields, which in turn may lead to technologies that are useful in our daily lives.
The key words of this lecture are "machine learning", "properties of semiconductors", and "galaxies". Please enjoy envisioning how science will open up our world in the near future from the respective fields of information science, physics, and astronomy.
Date, Time, and Program
Date: Friday, March 8, 2024, 14:00-17:00 (doors open 13:30)
Program:
14:00-14:05 Greetings from Dean of the School of Science
OHKOSHI Shin-ichi, Professor, Department of Chemistry
14:05-14:45 Lecture 1: "How Far Can Machine Learning Performance Improve?"
ISHIDA Takashi, Lecturer,
Department of Information Science (UG), Graduate School of Frontier Sciences
14:45-14:55 Q&A for Lecture 1
14:55-15:05 - Break
15:05-15:45 Lecture 2 "To be conductive, or not to be, that is the question."
AKIYAMA Ryota, Assistant Professor,
Department of Physics (GR), Graduate School of Science
15:45-15:55 Q&A for Lecture 2
15:55-16:05 - Break
16:05-16:45 Lecture 3 "The world of galaxies"
SHIMASAKU Kazuhiro, Associate Professor,
Department of Astronomy (GR), Graduate School of Science
16:45-16:55 Q&A for Lecture 3
17:00 End
Introduction of Lecturers and Lectures
Lecture 1 "How Far Can Machine Learning Performance Improve?"
ISHIDA Takashi, Lecturer, Department of Computer Science (GR), Graduate School of Information Science and Technology
Career Summary
Takashi Ishida is a Research Scientist with the Imperfect Information Learning Team at RIKEN AIP and a Lecturer at The University of Tokyo. He earned his PhD from the University of Tokyo in 2021.
Lecture Contents
Machine learning technology is now influencing every field, from our everyday lives to industry and scientific research, producing results beyond our imagination. However, what are the limits of machine learning performance? Within the framework of supervised learning, the concept of "Bayes error" provides an answer to this question. Bayes error represents the theoretically smallest possible error in classification problems and is an essential metric for evaluating the performance of machine learning models. In this talk, I will introduce methods to estimate Bayes error and its applications, based on our latest research. In the latter part of the talk, I will explore the potential future developments in the field of machine learning, discussing how far it may evolve and in what ways.
Lecture 2: "To be conductive, or not to be, that is the question."
AKIYAMA Ryota, Assistant Professor, Department of Physics (GR), Graduate School of Science
Career Summary
Ph. D. in 2012 from the Department of Electrical Engineering and Information Systems, Graduate School of Engineering, The University of Tokyo. Then, he was appointed as Assistant Professor in the Institute of Materials Science, Graduate School of Pure and Applied Sciences, University of Tsukuba in the same year. He has been in the current position since 2014. Ph. D. (Enginnering)
Lecture Contents
Metals conduct electricity as a matter of course. But wood and plastic do not conduct electricity. Then, in what conditions do materials conduct electricity? And what are semiconductors, which do not fall into these categories and are said to conduct electricity "sometimes"? The semiconductor, the miraculous stone that revolutionized the world in the 20th century, is now a material with a vastly expanded range of applications. The emergence of graphene, a material as thin as one atom, has led to the high-speed movement of electrons while confining them to two dimensions, and topological insulators, in which only the surface of the material is a conductor, have caused a variety of tectonic shifts. Focusing on semiconductors in the broad sense of the word, I will talk about the properties of these mysterious materials and how they will be utilized in society, while introducing the forefront of research.
Lecture 3 "The world of galaxies"
SHIMASAKU Kazuhiro, Associate Professor, Department of Astronomy (GR), Graduate School of Science
Career Summary
Doctor in Science, The University of Tokyo. Graduated from the Department of Astronomy, Graduate School of Science, The University of Tokyo (M.Sc). Present position from 2006 after working as assistant professor. Visiting researcher at Princeton University (1994).
Lecture Contents
In this talk, I will take you to the world of galaxies. What are galaxies? How do they evolve? Do supermassive black holes affect galaxy evolution!?
What is a hidden connection between galaxies and life? I will talk about these topics, showing the latest observations. No prior knowledge of galaxies is required.
Admission
Free of charge
How to register (on-site participation only, advance registration required)
Registration has closed as we have reached the maximum number of on-site participants (150). Thank you for all the applications.
Notes
- At the registration desk, we will confirm the e-mail automatically sent upon receipt of your application. Please note that if we cannot confirm your e-mail, you will not be allowed to enter the venue.
- Each lecture will be followed by a Q&A session. Questions will basically be accepted via Slido, so please bring your own device. Please enter the access code provided on the lecture screen or in the YouTube summary section into the Slido website to ask questions. Please note that the access code will be different for each presentation.
Location
Koshiba Hall, School of Science Building No. 1, Hongo Campus, The University of Tokyo Access Map
Live Streaming
Anyone can watch the lectures on the YouTube channel of the School of Science without prior registration. The lectures will be available on demand until 23:59 on March 10 (Sun.).
Other Information
■Materials for participants (PDF) will be posted on this page by the day before the event.
The materials will not be distributed at the venue, so on-site attendees are requested to download and read them here.
■School of Science, The University of Tokyo provides "Barrier-Free Support. If you need facilities or information security due to a disability, etc., please let us know at the time of application.
Organized by
School of Science, The University of Tokyo
Inquiries
Office of Communication, School of Science, The University of Tokyo
TEL: 03-5841-7585
email: kouhou.s@gs.mail.u-tokyo.ac.jp