Name
Abstract | ||
|---|---|---|
•Common and label-specific dynamical structures are extracted from sequential data.•We incorporate label data into dynamic mode decomposition via multitask learning.•We estimate label-wise sparse weights of dynamic modes by solving sparse-group Lasso.•We investigate the empirical performance using synthetic and real-world datasets.•Our methods show better accuracy than conventional dimensionality reduction methods. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1016/j.patrec.2019.02.010 | Pattern Recognition Letters |
Keywords | Field | DocType |
41A05,41A10,65D05,65D17 | Dynamic mode decomposition,Dimensionality reduction,Multi-task learning,Pattern recognition,Lasso (statistics),Dynamical systems theory,Nonlinear dynamical systems,Artificial intelligence,Mathematics,Modal | Journal |
Volume | ISSN | Citations |
122 | 0167-8655 | 0 |
PageRank | References | Authors |
0.34 | 13 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Keisuke Fujii | 1 | 2 | 1.39 |
| Kawahara, Yoshinobu | 2 | 317 | 31.30 |