Timezone: »
A spectral analysis of the Koopman operator, which is an infinite dimensional linear operator on an observable, gives a (modal) description of the global behavior of a nonlinear dynamical system without any explicit prior knowledge of its governing equations. In this paper, we consider a spectral analysis of the Koopman operator in a reproducing kernel Hilbert space (RKHS). We propose a modal decomposition algorithm to perform the analysis using finite-length data sequences generated from a nonlinear system. The algorithm is in essence reduced to the calculation of a set of orthogonal bases for the Krylov matrix in RKHS and the eigendecomposition of the projection of the Koopman operator onto the subspace spanned by the bases. The algorithm returns a decomposition of the dynamics into a finite number of modes, and thus it can be thought of as a feature extraction procedure for a nonlinear dynamical system. Therefore, we further consider applications in machine learning using extracted features with the presented analysis. We illustrate the method on the applications using synthetic and real-world data.
Author Information
Yoshinobu Kawahara (Osaka University)
More from the Same Authors
-
2018 Poster: Metric on Nonlinear Dynamical Systems with Perron-Frobenius Operators »
Isao Ishikawa · Keisuke Fujii · Masahiro Ikeda · Yuka Hashimoto · Yoshinobu Kawahara -
2017 Poster: Learning Koopman Invariant Subspaces for Dynamic Mode Decomposition »
Naoya Takeishi · Yoshinobu Kawahara · Takehisa Yairi -
2012 Poster: Weighted Likelihood Policy Search with Model Selection »
Tsuyoshi Ueno · Yoshinobu Kawahara · Kohei Hayashi · Takashi Washio -
2011 Poster: Prismatic Algorithm for Discrete D.C. Programming Problem »
Yoshinobu Kawahara · Takashi Washio -
2010 Spotlight: Minimum Average Cost Clustering »
Kiyohito Nagano · Yoshinobu Kawahara · Satoru Iwata -
2010 Poster: Minimum Average Cost Clustering »
Kiyohito Nagano · Yoshinobu Kawahara · Satoru Iwata -
2009 Poster: Submodularity Cuts and Applications »
Yoshinobu Kawahara · Kiyohito Nagano · Koji Tsuda · Jeffrey A Bilmes -
2009 Spotlight: Submodularity Cuts and Applications »
Yoshinobu Kawahara · Kiyohito Nagano · Koji Tsuda · Jeffrey A Bilmes -
2006 Poster: A Kernel Subspace Method by Stochastic Realization for Learning Nonlinear Dynamical Systems »
Yoshinobu Kawahara · Takehisa Yairi · Kazuo Machida