Neural networks

A Neural-Preconditioned Poisson Solver for Mixed Dirichlet and Neumann Boundary Conditions

We introduce a neural-preconditioned iterative solver for Poisson equations with mixed boundary conditions.

Topology Optimization via Frequency Tuning of Neural Design Representations

By tuning the scale hyperparameter of a neural design representation based on a Fourier feature network, we can control the smoothness of designs without explicit smoothing filters.