第3回 数理モデリング研究会 in 東京:
Workshop on multitrack event-trains in neural, social, seismological, and financial data

Date : 9-10, November, 2016.

Venue : National Institute of Informatics (NII)　 Room 1901-1903

Organizers:

Invited Speaker:

Dr. Patricia Reynaud-Bouret (Nice Sophia-Antipolis University) [www]

Aim:

The workshop is aimed at bringing together researchers working on point process models and analysis of multitrack event-trains in various fields in science, such as in neural, social, seismological, and financial data and developing the analytical tools and data-mining methods by sharing the ideas in various fields.

Program:

First day (11/9)
13:00 - 13:05 Takaaki Aoki
Opening address

13:05 - 14:05 Taro Takaguchi (National Institute of Information and Communications Technology) [www]
Classification of randomized reference models for temporal network data

14:35 - 15:35 Heetae Kim (Asia Pacific Centre for Theoretical Physics (APCTP), Korea) [www]
Dynamic Behaviour of Power Grids

16:00 - 17:30 (Invited) Patricia Reynaud-Bouret (Nice Sophia-Antipolis University) [www]
Detection of dependence between neurons and synchronization
Abstract Synchronization is an important phenomenon in neuroscience and some people think that this kind of dependency between neurons activities may play an important role in the neural code, that is how the brain codes information. But before entering this debate, there is another more mathematical /statistical question to answer: how do we detect such phenomenon in a set-up where the observations are scarce, noisy and always fluctuating ? I will start by describing Unitary Events methods (first introduced by Grün in the 90's) which makes rigorous the distinction between synchronization and pure coincidence and how statistical testing procedure can distinguish both. Then I will explain how to mimic distributions under independence to find out detection threshold for synchronization.


18:00 Dinner


Second day (11/10)
9:00 - 10:00 Takayuki Mizuno (National Institute of Informatics) [www]
Statistical laws and a stochastic model in online product markets

10:30 - 11:30 Tomoharu Iwata (NTT Communication Science Laboratories) [www]
Discovering Latent Influence via Shared Cascade Poisson Processes

12:00 - 13:00 Tsutomu Watanabe (University of Tokyo) [www]
Sticky Prices and Macroeconomic Fluctuations

13:00 - 14:30 Lunch

14:30 - 16:30
(Special session) Round table on the appplication of point process models to neural, social, seismological, and financial data

16:30 - 16:35 Ryota Kobayashi
Closing address

Photos:

Group Photo
この研究会は平成２８年度国立情報学研究所公募型共同研究「イベント時系列データマイニングの理論とその応用: Web•ソーシャルメディアデータから経済•脳•地震データ分析まで」の補助を受けています．