HCNN is a special type of recurrent neural network with teacher forcing while training. More detailed description is provided in https://link.springer.com/chapter/10.1007/978-3-642-28696-4_10
This package implements Historical Consistent Neural Network (HCNN) that was designed by Hans-Georg Zimmermann and has been studied by Alexey Minin in papers:
- Complex Valued Recurrent Neural Network: From Architecture to Training, https://www.researchgate.net/publication/273745545_Complex_Valued_Recurrent_Neural_Network_From_Architecture_to_Training
- Complex-valued open recurrent neural network for power transformer modeling, https://www.researchgate.net/publication/310516306_Complex-valued_open_recurrent_neural_network_for_power_transformer_modeling
Package also contains implementaion of Causal-Retro-Causal Neural Networks (CRC). Details are provided in https://link.springer.com/chapter/10.1007/978-3-642-29210-1_92.