A Physics-constrained super-resolution framework
Context: The aim is to proficiently depict high-resolution outputs and facilitate the integration of physical constraints, expressed as Partial Differential Equations (PDEs).
Proposed solution: MeshfreeFlowNet consists of two end-to-end trainable sub-networks, namely the Context Generation Network \(\Phi_{\theta_1}\) and the Continuous Decoding Network \(\Phi_{\theta_2}\), of parameters \(\theta_1\) and \(\theta_2\), respectively. The former produces a Latent Context Grid, by using a variant of the U-Net architecture, from a low-resolution physical input. Conversely, the latter generates physical quantities by means of a Multilayer Perceptron taking as input the concatenation of some context vector and a spatio-temporal coordinate. In order to generate physical quantities \(y_i\) at some high-scale coordinate \(x_i\), a trilinear interpolation of the output of the decoder is made :
\[\begin{equation} y_i =\sum_{j \in \mathcal{N}_i} w_j \Phi_{\theta_2}\left(\frac{x_i-x_j}{\Delta x}, c_j\right) \end{equation}\]where \(\mathcal{N}_i\) is the set of neighboring vertices that bound \(x_i\), \((x_j,c_j)\) are the spatio-temporal coordinates and the latent context vector for the \(j\)-th vertex of the low-resolution grid, and the \(w_j\)’s are interpolation weights.
The parameters of the sub-networks are learned by minimizing a weighted combination of the PDE residual \(\mathcal{L}_e\) and the prediction error \(\mathcal{L}_p\), both evaluated at the high-scale resolution.