Recurrent neural networks serve as black-box models for nonlinear dynamical systems identification and time series prediction. Training of recurrent networks typically minimizes t...
— In this work, a probabilistic model is established for recurrent networks. The EM (expectation-maximization) algorithm is then applied to derive a new fast training algorithm f...
We introduce a recurrent architecture having a modular structure and we formulate a training procedure based on the EM algorithm. The resulting model has similarities to hidden Ma...
An efficient algorithm to train general differential recurrent neural networks is proposed. The trained network can be directly used as the internal model of a predictive controll...
To predict the 100 missing values from a time series of 5000 data points, given for the IJCNN 2004 time series prediction competition, recurrent neural networks (RNNs) are trained...
Xindi Cai, Nian Zhang, Ganesh K. Venayagamoorthy, ...