Self-consistent recurrent neural network for path dependent deformation.

As an overall trend, the GRU architecture exhibited performance similar to the LSTM models, while in several cases achieving slightly lower validation mean squared error values.

also in 4,871 other papers