Presentation | 2018-11-05 [Poster Presentation] Discovery of Governing equations in Reproducing Kernel Hilbert space Yosuke Otsubo, Shinichi Nakajima, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | It has great value in lots of scientific fields, e.g., physics, chemistry, and biology, to discover governing equations through data-driven approaches. Dynamics which describe various phenomena can be modeled as nonlinear ordinary differential equations (ODEs). Although there are lots of studies for estimating the unknown parameters in ODEs, there are less the approaches for ODE discovery problem. We propose a method to address such a problem, based on sparse inference in the reproducing kernel Hilbert space. In our approach, we prepare candidates of possible bases, and then optimize the objective function consisting of two factors, one for data fidelity term for kernel ridge regression, and one for gradient matching with sparsity inducing regularizer. We validate the performance of our proposal through simulation studies in benchmark ODE systems. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Governing equations / Ordinary Differential Equations / Sparse inference / Reproducing Kernel Hilbert Space |
Paper # | IBISML2018-65 |
Date of Issue | 2018-10-29 (IBISML) |
Conference Information | |
Committee | IBISML |
---|---|
Conference Date | 2018/11/5(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Hokkaido Citizens Activites Center (Kaderu 2.7) |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Information-Based Induction Science Workshop (IBIS2018) |
Chair | Hisashi Kashima(Kyoto Univ.) |
Vice Chair | Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo) |
Secretary | Masashi Sugiyama(Nagoya Inst. of Tech.) / Koji Tsuda(AIST) |
Assistant | Tomoharu Iwata(NTT) / Shigeyuki Oba(Kyoto Univ.) |
Paper Information | |
Registration To | Technical Committee on Infomation-Based Induction Sciences and Machine Learning |
---|---|
Language | ENG-JTITLE |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Poster Presentation] Discovery of Governing equations in Reproducing Kernel Hilbert space |
Sub Title (in English) | |
Keyword(1) | Governing equations |
Keyword(2) | Ordinary Differential Equations |
Keyword(3) | Sparse inference |
Keyword(4) | Reproducing Kernel Hilbert Space |
1st Author's Name | Yosuke Otsubo |
1st Author's Affiliation | Nikon Corporation(Nikon) |
2nd Author's Name | Shinichi Nakajima |
2nd Author's Affiliation | Technical University of Berlin/Center for Advanced Intelligence Project (AIP), RIKEN(TUB/RIKEN) |
Date | 2018-11-05 |
Paper # | IBISML2018-65 |
Volume (vol) | vol.118 |
Number (no) | IBISML-284 |
Page | pp.pp.159-166(IBISML), |
#Pages | 8 |
Date of Issue | 2018-10-29 (IBISML) |