Learning with Coherence Patterns in Multivariate Time-series Data via Dynamic Mode Decomposition

التفاصيل البيبلوغرافية
العنوان: Learning with Coherence Patterns in Multivariate Time-series Data via Dynamic Mode Decomposition
المؤلفون: Masashi Hiraoka, Yoshinobu Kawahara, Takehito Bito
المصدر: IJCNN
بيانات النشر: IEEE, 2019.
سنة النشر: 2019
مصطلحات موضوعية: Multivariate statistics, Computer science, 02 engineering and technology, Coherence (statistics), 010501 environmental sciences, 01 natural sciences, Linear subspace, Set (abstract data type), Nonlinear system, Complex dynamics, Modal, 0202 electrical engineering, electronic engineering, information engineering, Dynamic mode decomposition, Embedding, 020201 artificial intelligence & image processing, Time series, Algorithm, 0105 earth and related environmental sciences
الوصف: Understanding complex dynamics in the real world is a fundamental problem in various engineering and scientific fields. Dynamic mode decomposition (DMD) has attracted attention recently as a prominent way to obtain global modal descriptions of nonlinear dynamical processes from data without requiring explicit prior knowledge about the underlying systems. In this paper, we propose a novel learning method for multivariate time-series data involving complex dynamics using coherence patterns among attributes extracted by DMD. To this end, we develop kernels defined with Grassmann subspaces spanned by dynamic modes which are calculated by DMD and represent coherence patters among attributes with respect to the estimated modal dynamics. To incorporate information in labels attached to a set of time-series sequences, we employ a supervised embedding step in the DMD procedure. We illustrate and investigate the empirical performance of the proposed method using real-world data.
DOI: 10.1109/ijcnn.2019.8852177
URL الوصول: https://explore.openaire.eu/search/publication?articleId=doi_________::253c6e7eb12ff2ce0204cf85c945376a
https://doi.org/10.1109/ijcnn.2019.8852177
Rights: CLOSED
رقم الانضمام: edsair.doi...........253c6e7eb12ff2ce0204cf85c945376a
قاعدة البيانات: OpenAIRE
الوصف
DOI:10.1109/ijcnn.2019.8852177