A hybrid model based on deep LSTM for predicting high-dimensional chaotic systems
This work addresses prediction challenges in chaotic systems, which is incremental as it builds on existing LSTM and empirical modeling approaches.
The authors tackled the problem of predicting high-dimensional chaotic systems by proposing a hybrid method combining deep LSTM with an empirical model, resulting in improved stability and reduced divergence compared to multi-layer LSTM alone, as shown by statistical metrics like RMSE and ACC on systems such as Mackey-Glass and Kuramoto-Sivashinsky.
We propose a hybrid method combining the deep long short-term memory (LSTM) model with the inexact empirical model of dynamical systems to predict high-dimensional chaotic systems. The deep hierarchy is encoded into the LSTM by superimposing multiple recurrent neural network layers and the hybrid model is trained with the Adam optimization algorithm. The statistical results of the Mackey-Glass system and the Kuramoto-Sivashinsky system are obtained under the criteria of root mean square error (RMSE) and anomaly correlation coefficient (ACC) using the singe-layer LSTM, the multi-layer LSTM, and the corresponding hybrid method, respectively. The numerical results show that the proposed method can effectively avoid the rapid divergence of the multi-layer LSTM model when reconstructing chaotic attractors, and demonstrate the feasibility of the combination of deep learning based on the gradient descent method and the empirical model.