A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems

•The present method can learn both linear and nonlinear correlations between the low- and high-fidelity data adaptively.•The present method can infer the quantities of interest based on a few scattered data.•The present method can identify the unknown parameters in the PDEs.•The present method can b...

Ausführliche Beschreibung