Deep learning prediction of amplitude death

Abstract Affected by parameter drift and coupling organization, nonlinear dynamical systems exhibit suppressed oscillations. This phenomenon is called amplitude death. In various complex systems, death a typical critical phenomenon, which may lead to the functional collapse of system. Therefore, an important issue how effectively predict phenomena based on data in system oscillation state. paper proposes enhanced Informer model The employs attention mechanism capture long-range associations time series tracks effect dynamics through accompanying input channel. experimental results coupled Rössler Lorentz show that informer has higher prediction accuracy longer effective distance than original algorithm can