| |  Online-Ressource |
|---|
| Verfasst von: | Schmidt, Dominik [VerfasserIn]  |
|---|
| | Koppe, Georgia [VerfasserIn] |
|---|
| | Monfared, Zahra [VerfasserIn] |
|---|
| | Beutelspacher, Max [VerfasserIn] |
|---|
| | Durstewitz, Daniel [VerfasserIn] |
|---|
| Titel: | Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies |
|---|
| Verf.angabe: | Dominik Schmidt, Georgia Koppe, Zahra Monfared, Max Beutelspacher, Daniel Durstewitz |
|---|
| Ausgabe: | Version v3 |
|---|
| E-Jahr: | 2021 |
|---|
| Jahr: | 12 Mar 2021 |
|---|
| Umfang: | 29 S. |
|---|
| Illustrationen: | Illustrationen |
|---|
| Fussnoten: | Online veröffentlicht am 8. Oktober 2019, Version 2 am 19. Juni 2020, Version 3 am 12. März 2021 ; Gesehen am 10.01.2024 |
|---|
| Titel Quelle: | Enthalten in: Arxiv |
|---|
| Ort Quelle: | Ithaca, NY : Cornell University, 1991 |
|---|
| Jahr Quelle: | 2021 |
|---|
| Band/Heft Quelle: | (2021), Artikel-ID 1910.03471, Seite 1-29 |
|---|
| Abstract: | A main theoretical interest in biology and physics is to identify the nonlinear dynamical system (DS) that generated observed time series. Recurrent Neural Networks (RNNs) are, in principle, powerful enough to approximate any underlying DS, but in their vanilla form suffer from the exploding vs. vanishing gradients problem. Previous attempts to alleviate this problem resulted either in more complicated, mathematically less tractable RNN architectures, or strongly limited the dynamical expressiveness of the RNN. Here we address this issue by suggesting a simple regularization scheme for vanilla RNNs with ReLU activation which enables them to solve long-range dependency problems and express slow time scales, while retaining a simple mathematical structure which makes their DS properties partly analytically accessible. We prove two theorems that establish a tight connection between the regularized RNN dynamics and its gradients, illustrate on DS benchmarks that our regularization approach strongly eases the reconstruction of DS which harbor widely differing time scales, and show that our method is also en par with other long-range architectures like LSTMs on several tasks. |
|---|
| DOI: | doi:10.48550/arXiv.1910.03471 |
|---|
| URL: | Bitte beachten Sie: Dies ist ein Bibliographieeintrag. Ein Volltextzugriff für Mitglieder der Universität besteht hier nur, falls für die entsprechende Zeitschrift/den entsprechenden Sammelband ein Abonnement besteht oder es sich um einen OpenAccess-Titel handelt.
kostenfrei: Volltext: https://doi.org/10.48550/arXiv.1910.03471 |
|---|
| | kostenfrei: Volltext: http://arxiv.org/abs/1910.03471 |
|---|
| | DOI: https://doi.org/10.48550/arXiv.1910.03471 |
|---|
| Datenträger: | Online-Ressource |
|---|
| Sprache: | eng |
|---|
| Sach-SW: | Computer Science - Machine Learning |
|---|
| | Quantitative Biology - Quantitative Methods |
|---|
| | Statistics - Machine Learning |
|---|
| K10plus-PPN: | 1818272873 |
|---|
| Verknüpfungen: | → Sammelwerk |
|---|
Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies / Schmidt, Dominik [VerfasserIn]; 12 Mar 2021 (Online-Ressource)
