Volume 89
Back to Newsletter List
Volume 89
February 14, 2020
Seminar Report
Math Seminar Talk by Dr. Martin Skrodzki, February 7, 2020.
2020-02-12
Dr. Martin Skrodzki, who stays at RIKEN iTHEMS as a postdoctoral researcher for the year 2020, gave a talk at the Math Seminar on February 7, 2020. The title of his talk was "Solved and open problems regarding the neighborhood grid data structure". He began with introducing the k-d tree for a finite set with coordinates in the plane and explained that by using this tree one can find the nearest point from a given point in a reasonable time. Then, he looked at the neighborhood grid data structure introduced by Joselli et al. in 2009, which is an n times n matrix filled by pairs of numbers (f(i,j),g(i,j)) where f and g are surjections to the set of numbers from 1 to n squared. He gave the definition of a stable state and explained that any grid data structure can be stabilized and that stabilization can be done in a short time using parallel computation. He explained the correspondence between a generic set of n squared points in the plane and stable grid data structures, and discussed several open questions around this correspondence. The talk was very accessible for everyone and the participants enjoyed it very much.
Solved and open problems regarding the neighborhood grid data structure
February 7 (Fri) at 16:00 - 18:10, 2020
Seminar Report
Biology Talk by Dr. Yasuo Yasui, February 5, 2020.
2020-02-13
On February 5, 2020, Dr. Yasuo Yasui, a plant geneticist from Kyoto University, gave a talk on his research on buckwheat (=soba) genetics. First, he explained how the current food supply of the world is heavily dependent on a very small number of plants, and that there is a pressing need to increase the yield of many other non-major crops, including buckwheat. He argued that now we have the tools to tackle this problem thanks to the advance in genome sequencing technologies, and that data science, computer science, and mathematical science have important roles to play. He then presented his previous research on buckwheat genetics and genomics, such as the identification of certain genes in buckwheat that are important for buckwheat breeding. He also showed some slides from field trips in South China (mainly Yunnan province) to collect wild buckwheat species, and explained his ongoing research in trying to understand the origin and domestication process of buckwheat, which is a joint effort with Jeffrey Fawcett from iTHEMS, and other researchers in Japan, China, and the UK. The talk was aimed at non-biologists, and many non-biologists from iTHEMS were able to join and ask many questions.
Biology Talk
February 5 (Wed) at 16:00 - 16:30, 2020
Upcoming Events
Seminar
Mining for Dark Matter substructure: Learning from lenses without a likelihood
February 17 (Mon) at 14:00 - 15:30, 2020
Johann Brehmer (Postdoctoral Researcher, New York University, USA)
Dr. Brehmer gives us a talk about a method to deduce DM small structures. Please join us!
The subtle imprint of dark matter substructure on extended arcs in strong lensing systems contains a wealth of information about the small-scale distribution of dark matter and, consequently, about the underlying particle physics. However, teasing out this effect is challenging since the likelihood function for realistic simulations of population-level parameters is intractable. Structurally similar problems appear in many other scientific fields ranging from particle physics to neuroscience to epidemiology, which has prompted the development of powerful simulation-based inference techniques based on machine learning. We give a broad overview over these methods, and then apply them to the problem of substructure inference in galaxy-galaxy strong lenses. In this proof-of-principle application to simulated data, we show that these methods can provide an efficient and principled way to simultaneously analyze an ensemble of strong lenses, and can be used to mine the large sample of lensing images deliverable by near-future surveys for signatures of dark matter substructure.
Venue: #424-426, Main Research Building, RIKEN
Event Official Language: English
Colloquium
The 11st MACS Colloquium
February 19 (Wed) at 15:00 - 17:30, 2020
Hiroyuki Hata (Professor, Graduate School of Science, Kyoto University)
15:00- Teatime
15:15- Talk by Prof. Hiroyuki Hata
16:40- MACS Report Meeting FY2019
The 11th MACS colloquium is supported by iTHEMS. It will be broadcasted to Wako, but if you can join the colloquium physically in Kyoto, that would be better. iTHEMS provides good confectionery at Kyoto!
Venue: Lecture room #401, Graduate School of Science Building No 6, Kyoto University
Event Official Language: Japanese
Seminar
Index of the Wilson-Dirac operator revisited: a discrete version of Dirac operator on a finite lattice
February 25 (Tue) at 16:00 - 18:10, 2020
Mikio Furuta (Professor, The University of Tokyo)
The Wilson-Dirac operator is a discrete version of Dirac operator defined on regular lattices. When the discrete version is a fine approximation of the Dirac operator on a Z/2-graded Clifford module on a torus, it is known that (1) an integer-valued index is defined for the Wilson-Dirac operator, and (2) the index is equal to the Atiyah-Singer index of the Dirac operator on the torus.
These have been well established up to around 2000. The strategy of all the previous works is to make use of the discrete version of the heat kernel for Neuberger's overlap Dirac operator. Therefore the strategy cannot be generalized to mod 2 index nor family version of index.
In this talk I would like to explain a new approach to the index of Wilson-Dirac operator which can be immediately generalized to these various cases.
Joint work with H. Fukaya, S. Matsuo, T. Onogi, S. Yamaguchi and M. Yamashita.
Venue: Seminar Room #160, 1F Main Research Building, RIKEN
Event Official Language: English
If you would like to cancel your subscription or change your email address,
please let us know via our contact form.
Copyright © iTHEMS, RIKEN. All rights reserved.