Scalable physical source-to-field inference with hypernetworks

We present a generative model that amortises computation for the field around e.g. gravitational or magnetic sources. Exact numerical calculation has either computational complexity in the number of sources and field evaluation points, or requires a fixed evaluation grid to exploit fast Fourier transforms. Using an architecture where a hypernetwork produces an implicit representation of the field around a source collection, our model instead performs as , achieves accuracy of , and allows evaluation at arbitrary locations for arbitrary numbers of sources, greatly increasing the speed of e.g. physics simulations. We also examine a model relating to the physical properties of the output field and develop two-dimensional examples to demonstrate its application. The code for these models and experiments is available at https://github.com/cmt-dtu-energy/hypermagnetics.
View on arXiv