Fig. 11From: Non-intrusive nonlinear model reduction via machine learning approximations to low-dimensional operatorsPareto frontier of relative error with respect to the relative running time using 4th-order Runge–Kutta for 2D convection–diffusion equation: a \({{e}}_\textit{FOM}\) vs. \({\varvec{\tau }}_\textit{FOM}\) in FOM; b \({{e}}_\textit{ROM}\) vs. \({\varvec{\tau }}_\textit{ROM}\) in ROMBack to article page