FiniteNet: A Fully Convolutional LSTM Network Architecture for Time-Dependent Partial Differential Equations

1 code implementation • 7 Feb 2020 • Ben Stevens, Tim Colonius

In this work, we present a machine learning approach for reducing the error when numerically solving time-dependent partial differential equations (PDE).

Enhancement of shock-capturing methods via machine learning

no code implementations • 6 Feb 2020 • Ben Stevens, Tim Colonius

Because data can be used to learn nonlinear relationships, we train a neural network to improve the results of a fifth-order WENO method.

BIG-bench Machine Learning