Physics-informed Neural Networks (PINNs)
Physics-informed machine learning allows scientists to use this prior knowledge to help the training of the neural network, making it more efficient. This means it will need fewer samples to train it well or to make the training more accurate.
PINNs are used to estimate the results for a given physical system consisting of a PDE, initial conditions (ICs), and BCs by minimizing constraints in the loss function. PINNs utilize the automatic differentiation of deep learning frameworks to compute the derivatives of the PDE to compute the residual error.