A GNN-based surrogate model predicts MoM-quality surface currents on 3D conducting bodies, trading 2-3x accuracy for 3-5x faster training compared to PhiGRL.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Study on a Fast Solver for Combined Field Integral Equations of 3D Conducting Bodies Based on Graph Neural Networks
A GNN-based surrogate model predicts MoM-quality surface currents on 3D conducting bodies, trading 2-3x accuracy for 3-5x faster training compared to PhiGRL.