Machine Learning for Auxiliary Sources
This work provides an incremental advancement for computational electromagnetics researchers by reframing an existing numerical method within a neural network framework.
The authors reformulate the Method of Auxiliary Sources (MAS), a numerical technique for Partial Differential Equations, as a neural network. This allows them to use optimization algorithms like Adam to train MAS and determine optimal coefficients and positions of source singularities. They also demonstrate that this neural network-trained MAS can detect the position of an unknown function's central singularity.
We rewrite the numerical ansatz of the Method of Auxiliary Sources (MAS), typically used in computational electromagnetics, as a neural network, i.e. as a composed function of linear and activation layers. MAS is a numerical method for Partial Differential Equations (PDEs) that employs point sources, which are also exact solutions of the considered PDE, as radial basis functions to match a given boundary condition. In the framework of neural networks we rely on optimization algorithms such as Adam to train MAS and find both its optimal coefficients and positions of the central singularities of the sources. In this work we also show that the MAS ansatz trained as a neural network can be used, in the case of an unknown function with a central singularity, to detect the position of such singularity.