Regensburg 2019 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 63: Modeling and Data Analysis

DY 63.4: Talk

Friday, April 5, 2019, 10:45–11:00, H20

DyCA - Dynamical Component Analysis — Steffen Hartmann, •Katharina Korn, Bastian Seifert, and Christian Uhl — Center for Signal Analysis of Complex Systems, University of Applied Sciences Ansbach, Residenzstr. 8, 91522 Ansbach

Spatio-temporal patterns emerging from a certain class of low-dimensional differential equations can be decomposed by dynamical component analysis (DyCA), a new algorithm proposed in [1]. We present the method, which is based on a least-square-optimization leading to a generalized eigenvalue problem of correlation matrices of the signal and its derivative. The application of DyCA to simulated and real world data is demonstrated and the power of the algorithm is shown by comparing the results to principal and independent component analysis (PCA, ICA).

[1] B. Seifert, K. Korn, S. Hartmann, and C. Uhl, Dynamical Component Analysis (DyCA): Dimensionality reduction for high-dimensional deterministic time-series, 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP), Aalborg, 17.-20.09.2018