DAIKIN International Symposium on Physics of Intelligence-- Statistical Mechanics and Machine Learning: A Powerful Combination for Data Analysis --

ISPI2024: Nov. 6-8, 2024 @Koshiba Hall, The University of Tokyo

【Access】

Koshiba Hall (No. 100 in Hongo Campus Map)
Access Map to Hongo Campus

【Objective】

The goal of machine learning is to extract underlying regularities from training data. The objective is no different from that of statistics, which has been developed for 200 years since Laplace. However, in machine learning, the probabilistic models used to extract regularities are nonlinear and have much higher degrees of freedom than previous statistical models. This leads to new challenges, such as the difficulty of computation and the difficulty of performance evaluation in the data analysis. On the other hand, statistical mechanics has greatly developed techniques for dealing with large-degree-of-freedom nonlinear statistical models through the study of gases and magnetic materials. This suggests that statistical mechanics may be useful for solving the new challenges posed by the birth of machine learning. This symposium aims to bring together researchers in data analysis, machine learning, and statistical mechanics to exchange their expertise.

[Registration] 

To be announced.

[Application to poster presentation]

To be announced.

【Speakers】                         

*plenary talk

※In alphabetical order
SueYeon Chung New York University (NYU) USA
Jorn Dunkel Massachusetts Institute of Technology (MIT) USA
Alexander Hoffmann University of California, Los Angeles (UCLA) USA
Sosuke Ito The University of Tokyo (UTokyo) JPN
Shinpei Kawaoka Tohoku University/Kyoto University JPN
Takeshi Kawasaki Nagoya University JPN
Nobuyasu Koga Osaka University JPN
Jian Ma Carnegie Mellon University (CMU) USA
Hiroshi Makino Nanyang Technological University (NTU) SGP
Marc Mézard * Università Bocconi ITA
Daiki Nishiguchi UTokyo JPN
Mor Nitzan Hebrew University of Jerusalem (HUJI) ISR
Mariko Okada Osaka University JPN
Cengiz Pehlevan Harvard University USA
Gautam Reddy Harvard University USA
Sunghan Ro MIT USA
Yasushi Sako Riken JPN
Kaoru Sugimura UTokyo JPN
Shinsuke Uda Yamaguchi University JPN
Vincenzo Vitelli The University of Chicago USA
Lei Wang Chinese Academy of Sciences (CAS) CHN
Matthiew Wyart École polytechnique fédérale de Lausanne (EPFL) CHE
Sho Yaida Meta USA
Hajime Yoshino Osaka Univeristy JPN
Francesco Zamponi Sapienza University ITA

【ISPI2024 Organizing Committee】

Yoshiyuki Kabashima
Kazumasa A. Takeuchi
Kyogo Kawaguchi

【Sponsors】

DAIKIN INDUSTRIES, LTD
Institute for Physics of Intelligence, The University of Tokyo
JST JPMJCR1912 "Deciphering intracellular phenomena through information flow”

【Contact】

ispi2024[at]ipi-ut.org

←to ipi TOP 
  • このエントリーをはてなブックマークに追加