Long Short-term Cognitive Network (LSTCN) model for time series forecasting
The Long Short-term Cognitive Network (LSTCN) model [1] is an efficient recurrent neural network for time series forecasting. It supports both one-step-ahead and multiple-step-ahead forecasting of univariate and multivariate time series. The LSTCN model is competitive compared to state-of-the-art recurrent neural networks such as LSTM and GRU in terms of forecasting error while being much faster.