<ul><li>Kolmogorov-Arnold Networks (KANs) are being used in deep learning applications in computational physics.</li><li>Physics-informed Kolmogorov-Arnold PointNet (PI-KAN-PointNet) allows simultaneous solution of an inverse problem over multiple irregular geometries in a single training run.</li><li>PI-KAN-PointNet provides more accurate predictions compared to physics-informed PointNet with MLPs.</li><li>Combining KAN and MLP in constructing a physics-informed PointNet leads to the optimal configuration.</li></ul>

Physics-informed KAN PointNet: Deep learning for simultaneous solutions to inverse problems in incompressible flow on numerous irregular geometries

Discover more