physics.geo-phNov 14, 2026

NAP-Wave: A Neural Adaptive Plane-Wave Architecture for High-Frequency Helmholtz Problems

Authors: Hamza Kamil, Mohammad Mahdi Abedi, David Pardo

Organizations: Basque Center for Applied Mathematics (BCAM), Bilbao, Spain · University of the Basque Country (UPV/EHU), Leioa, Spain · Ikerbasque: Basque Foundation for Science, Bilbao, Spain

Abstract

Training neural networks to simulate highly oscillatory wavefields presents optimization challenges, often hindered by spectral bias and poorly conditioned physics-based loss landscapes. To address these challenges and enhance convergence, we propose NAP-Wave, a joint neural architecture and training strategy that expands the target scattered wavefield in an adaptive neural plane-wave basis. The basis functions are extracted from all hidden layers and modulated by trainable plane-wave carriers, embedding the high-frequency oscillatory physics directly into the network's trial space. NAP-Wave is trained on a physics-informed loss built from the Lippmann-Schwinger integral equation, so no differentiation of the network with respect to its input is required. We develop a decoupled optimization strategy in which Adam updates the parameters of the neural plane-wave basis, while the linear combination coefficients are computed exactly by a least-squares (LS) solver. We couple this with an adaptive-depth training strategy in which the network starts shallow and grows progressively layer-by-layer. At each growth step, we avoid the loss spikes typical of dynamic network growth using the LS-computed coefficients, the warm-started parameters, and the preservation of the first- and second-order optimizer moments of the previous stage. Ablation studies confirm that the plane-wave basis, the exact coefficient solve, and the progressive schedule independently accelerate convergence and improve accuracy. Validated on exact-solution benchmarks, NAP-Wave achieves relative L2 errors of order 1%. On highly heterogeneous geophysical benchmarks, it achieves relative errors of 5%, 9%, and 10%, significantly outperforming published PINN methods in both accuracy and convergence time. To ensure reproducibility, the code is available at https://github.com/Hamza-Kamil/NAP-WAVE.

Figures & tables

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Gabor-Enhanced Physics-Informed Neural Networks for Fast Simulations of Acoustic Wavefields

    Feb 24, 2025Mohammad Mahdi Abedi, David Pardo, Tariq AlkhalifahNeural PDE SolversPDE Solving

  2. A Green-Integral-Constrained Neural Solver with Stochastic Physics-Informed Regularization

    Apr 23, 2026Mohammad Mahdi Abedi, David Pardo, Tariq AlkhalifahNeural PDE SolversPDE Solving

  3. Neural network-driven domain decomposition for efficient solutions to the Helmholtz equation

    Nov 19, 2025Victorita Dolean, Daria Hrebenshchykova, Stéphane Lanteri +1Neural PDE SolversPDE Solving