LOGIN ENGLISH

제목 [학술세미나] SEED 세미나 안내(18.10.31)
작성일 2018-10-26 13:06:56
내용 SEED Seminar Announcement


We cordially invite you to attend the SEED seminar by Prof. Shuhei Mano on Wednesday, 31 Oct 2018@17:00-18:00PM (UTC+9 note time difference).   
 
* Title: A Direct Sampler from Log-affine Models with Aid of Computational Algebra
* Speaker: Shuhei Mano, ISM
* Time: 17:00-18:00 (Japan time), Wednesday,  31 Oct  2018 
* Venue: ISM, Japan
* Virtual seminar roomwebconf.vc.dfn.de/optimization  
* Room Passcode: seed
 
Abstract:
Multinomial sampling from log-affine models is very common in count data analyses, including the two-by-two contingency table with fixed marginal sums. The Markov chain Monte Carlo (MCMC) is a very popular methods, because it does not need normalizing constants. However, it is well known that MCMC has several drawbacks, including departure from the stationarity and auto correlation among samples. Diaconis and Sturmfels (1998, Ann. Stat.) proposed use of the theory of Groebner basis to study the Markov bases, which are bases for an MCMC sampler. Their work is one of a origin of the recent developments of algebraic statistics. In this talk, I will introduce a direct sampler from log-affaine models (M 2017, Electron. J. Stat.), which enables independent sampling from the exact distribution. It is based on the Weyl algebra with the A-hypergeometric system, which was introduced by Israel Gel'fand and coauthors in late 1980s. In this talk, I will concentrate on the two-by-two contingency table, because every statistician knows about it. I will explain details why and how we can directly sample from the distribution with aid of computational algebra. This talk is partly based on a joint work with Nobuki Takayama at Kobe university.
 
 
-------------------------------------------------------------------------------------------------------------------------
The SEED seminar series is jointly organized by researchers from National University of Singapore, Zuse Institute Berlin, The Institute of Statistical Mathematics, Academia Sinica, University College London, Seoul National University and the Hong Kong University of Science and Technology. SEED stands for Statistics maschinElEarning Datascience. Motivated by the availability of big complex data and the fast development of new techniques in machine learning and data science, SEED aims to provide an online research platform for seminars focusing on important and timely interdisciplinary research topics on Statistics, Machine learning, Data Science, Mathematics, Operation Research, Computer Science, and Engineering. The online seminar series are co-hosted and organised by several research institutes in different countries. The mission is to exchange research ideas, educate young researchers, and promote international research and education collaborations.For more information, please visit the SEED website https://seed.stat.nus.edu.sg/index.php 
--------------------------------------------------------------------------------------------------------------------------
We look forward to seeing you at the events!
 
Best regards
 
SEED Organisation Committee
  
********************************************************************************************************
 
Confirmed seminars will be given by the following researchers:
 
Speaker Title Date and Time Host
Hong Kong, Singapore and Taiwan (UTC +8) Japan and Korea 
(UTC +9)
Germany 
(UTC +1)
U.K. 
(UTC)
Hideitsu Hino Current Dipole Localization from EEG with Birth-Death Process Wed, 9 Jan 2019
16:00 - 17:00
Wed, 9 Jan 2019
17:00 - 18:00
Wed, 9 Jan 2019
09:00 - 10:00
Wed, 9 Jan 2019
08:00 - 09:00
ISM
Guido Germano Integral transform methods and spectral filters for the pricing of exotic options Fri, 14 Dec 2018
17:30 - 18:30
Fri, 14 Dec 2018
18:30 - 19:30
Fri, 14 Dec2018
10:30 - 11:30
Fri, 14 Dec 2018
09:30 - 10:30
UCL
Simone Righi Social closure and the evolution of cooperation via indirect reciprocity Fri, 16 Nov 2018
18:00 - 19:00
Fri, 16 Nov 2018
19:00 - 20:00
Fri, 16 Nov2018
11:00 - 12:00
Fri, 16 Nov2018
10:00 - 11:00
UCL
  
 
파일 -