Critical transitions and qualitative changes of dynamics in cardiac, ecological, and economical systems, can be characterized by discrete-time bifurcations. The authors propose a deep learning framework that provides early warning signals for critical transitions in discrete-time experimental data.
- Thomas M. Bury
- Daniel Dylewsky
- Gil Bub
