Predicting discrete-time bifurcations with deep learning
© 2023. Springer Nature Limited..
Many natural and man-made systems are prone to critical transitions-abrupt and potentially devastating changes in dynamics. Deep learning classifiers can provide an early warning signal for critical transitions by learning generic features of bifurcations from large simulated training data sets. So far, classifiers have only been trained to predict continuous-time bifurcations, ignoring rich dynamics unique to discrete-time bifurcations. Here, we train a deep learning classifier to provide an early warning signal for the five local discrete-time bifurcations of codimension-one. We test the classifier on simulation data from discrete-time models used in physiology, economics and ecology, as well as experimental data of spontaneously beating chick-heart aggregates that undergo a period-doubling bifurcation. The classifier shows higher sensitivity and specificity than commonly used early warning signals under a wide range of noise intensities and rates of approach to the bifurcation. It also predicts the correct bifurcation in most cases, with particularly high accuracy for the period-doubling, Neimark-Sacker and fold bifurcations. Deep learning as a tool for bifurcation prediction is still in its nascence and has the potential to transform the way we monitor systems for critical transitions.
| Medienart: |
E-Artikel |
|---|
| Erscheinungsjahr: |
2023 |
|---|---|
| Erschienen: |
2023 |
| Enthalten in: |
Zur Gesamtaufnahme - volume:14 |
|---|---|
| Enthalten in: |
Nature communications - 14(2023), 1 vom: 10. Okt., Seite 6331 |
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Bury, Thomas M [VerfasserIn] |
|---|
| Links: |
|---|
| Themen: |
|---|
| Anmerkungen: |
Date Completed 01.11.2023 Date Revised 19.11.2023 published: Electronic Citation Status MEDLINE |
|---|
| doi: |
10.1038/s41467-023-42020-z |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
NLM363098240 |
|---|
| LEADER | 01000naa a22002652 4500 | ||
|---|---|---|---|
| 001 | NLM363098240 | ||
| 003 | DE-627 | ||
| 005 | 20231226092628.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 231226s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1038/s41467-023-42020-z |2 doi | |
| 028 | 5 | 2 | |a pubmed24n1210.xml |
| 035 | |a (DE-627)NLM363098240 | ||
| 035 | |a (NLM)37816722 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 100 | 1 | |a Bury, Thomas M |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Predicting discrete-time bifurcations with deep learning |
| 264 | 1 | |c 2023 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ƒaComputermedien |b c |2 rdamedia | ||
| 338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date Completed 01.11.2023 | ||
| 500 | |a Date Revised 19.11.2023 | ||
| 500 | |a published: Electronic | ||
| 500 | |a Citation Status MEDLINE | ||
| 520 | |a © 2023. Springer Nature Limited. | ||
| 520 | |a Many natural and man-made systems are prone to critical transitions-abrupt and potentially devastating changes in dynamics. Deep learning classifiers can provide an early warning signal for critical transitions by learning generic features of bifurcations from large simulated training data sets. So far, classifiers have only been trained to predict continuous-time bifurcations, ignoring rich dynamics unique to discrete-time bifurcations. Here, we train a deep learning classifier to provide an early warning signal for the five local discrete-time bifurcations of codimension-one. We test the classifier on simulation data from discrete-time models used in physiology, economics and ecology, as well as experimental data of spontaneously beating chick-heart aggregates that undergo a period-doubling bifurcation. The classifier shows higher sensitivity and specificity than commonly used early warning signals under a wide range of noise intensities and rates of approach to the bifurcation. It also predicts the correct bifurcation in most cases, with particularly high accuracy for the period-doubling, Neimark-Sacker and fold bifurcations. Deep learning as a tool for bifurcation prediction is still in its nascence and has the potential to transform the way we monitor systems for critical transitions | ||
| 650 | 4 | |a Journal Article | |
| 650 | 4 | |a Research Support, Non-U.S. Gov't | |
| 700 | 1 | |a Dylewsky, Daniel |e verfasserin |4 aut | |
| 700 | 1 | |a Bauch, Chris T |e verfasserin |4 aut | |
| 700 | 1 | |a Anand, Madhur |e verfasserin |4 aut | |
| 700 | 1 | |a Glass, Leon |e verfasserin |4 aut | |
| 700 | 1 | |a Shrier, Alvin |e verfasserin |4 aut | |
| 700 | 1 | |a Bub, Gil |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Nature communications |d 2010 |g 14(2023), 1 vom: 10. Okt., Seite 6331 |w (DE-627)NLM199274525 |x 2041-1723 |7 nnns |
| 773 | 1 | 8 | |g volume:14 |g year:2023 |g number:1 |g day:10 |g month:10 |g pages:6331 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1038/s41467-023-42020-z |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a GBV_NLM | ||
| 951 | |a AR | ||
| 952 | |d 14 |j 2023 |e 1 |b 10 |c 10 |h 6331 | ||