Inductively Scalable, Single-Step Neural Surrogates for Wave-Scattering Inverse Problems

• URL: http://arxiv.org/abs/2608.17344v1
• Date: Tue, 18 Aug 2026 04:07:53 GMT
• Title: Inductively Scalable, Single-Step Neural Surrogates for Wave-Scattering Inverse Problems
• Abstract summary: Neural network surrogates are an emerging alternative to traditional electromagnetic wave simulators.<n>We show that this barrier can be overcome by dynamically generating salient training examples during training.<n>We demonstrate the surrogate on large-scale forward simulations and inverse design of freeform beam splitters and gradientindex lenses up to 98 wavelengths wide.
• Score: 0.8164433158925593
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Neural network surrogates are an emerging alternative to traditional electromagnetic wave simulators like finite-difference time-domain (FDTD); their goal is to replace rigorous physical simulations with pre-trained neural networks that solve wave-scattering forward and inverse problems orders of magnitude faster. However, nonrecurrent, single-step surrogates have scaled only to a few tens of simulation variables. Here, we show that this barrier can be overcome by dynamically generating salient training examples during training, rather than randomly sampling the large space of possible examples. We introduce an algorithm that runs in parallel with surrogate training, using gradient ascent to search refractive-index and source configurations for cases where the surrogate disagrees with a full-wave ground-truth simulator. We also use source and ground-truth normalization with an evolving replay dataset to stabilize and accelerate learning. Using this approach, we train a fast, single-step surrogate for two-dimensional wave scattering with up to 41,772 controllable variables, including dense, freely configurable grids of refractive indices and complex-valued sources. The resulting neural surrogate is robustly accurate across diverse structured and unstructured examples and generalizes inductively to larger domains, reaching over 3 million controllable variables without retraining, a $73.8\times$ increase. We demonstrate the surrogate on large-scale forward simulations and inverse design of freeform beam splitters and gradient-index (GRIN) lenses up to 98 wavelengths wide, showing comparable or better performance than FDTD-based designs, with speedups from $1.29\times$ to $26.5\times$. These results demonstrate a practical path toward fast, robustly accurate, inductively scalable neural simulators for photonic inverse design and other wave-scattering inverse problems.

Related papers

• Taming Curvature: Architecture Warm-Up for Stable Transformer Training [52.6838951211596]
  We introduce a fast online estimator of the largest (preconditioned) Hessian eigenvalue (i.e., curvature) based on a warm-started variant for power with Hessian-vector products.<n>We find that training instabilities coincide with surges in preconditioned curvature and that curvature grows with depth.<n> Experiments on large Transformers validate that our approach enables efficient curvature tracking and reduces instabilities.
  arXiv  Detail & Related papers  (2026-06-15T14:16:56Z)
• A Decoupled Basis-Vector-Driven Generative Framework for Dynamic Multi-Objective Optimization [13.136405879524856]
  This paper proposes a basis-vector-driven generative framework (DB-GEN) for dynamic multi-objective optimization.<n>It addresses non-linear coupling, negative transfer from outdated historical data, and the cold-start problem during environmental switches.<n> Experimental results demonstrate that DB-GEN improves tracking accuracy across various dynamic benchmarks compared to existing algorithms.
  arXiv  Detail & Related papers  (2026-04-01T05:47:56Z)
• Ada3Drift: Adaptive Training-Time Drifting for One-Step 3D Visuomotor Robotic Manipulation [53.750389076941396]
  Diffusion-based visuomotor policies capture multimodal action distributions through iterative denoising, but their high inference latency limits real-time robotic control.<n>We propose Ada3Drift, which learns a training-time drifting field that attracts predicted actions toward expert demonstration modes.<n>Ada3Drift achieves state-of-the-art performance while requiring $10times$ fewer function evaluations than diffusion-based alternatives.
  arXiv  Detail & Related papers  (2026-03-12T14:36:53Z)
• Advancing accelerator virtual beam diagnostics through latent evolution modeling: an integrated solution to forward, inverse, tuning, and UQ problems [46.568487965999225]
  We propose Latent Evolution Model (LEM), a hybrid machine learning framework with an autoencoder that projects high-dimensional phase spaces into lower-dimensional representations, and transformers to learn temporal dynamics in the latent space.<n>For textitforward modeling, a CVAE encodes 15 unique projections of the 6D phase space into a latent representation, while a transformer predicts downstream latent states from upstream inputs.<n>For textitinverse problems, we address two distinct challenges: (a) predicting upstream phase spaces from downstream observations, and (b) estimating RF settings from the latent space of
  arXiv  Detail & Related papers  (2026-02-26T04:46:26Z)
• Is Flow Matching Just Trajectory Replay for Sequential Data? [46.770624059457724]
  Flow matching (FM) is increasingly used for time-series generation.<n>It is not well understood whether it learns a general dynamical structure or simply performs an effective "trajectory replay"<n>We show that the implied sampler is an ODE whose dynamics constitutes a nonparametric, memory-augmented continuous-time dynamical system.
  arXiv  Detail & Related papers  (2026-02-09T06:48:45Z)
• Physics Enhanced Deep Surrogates for the Phonon Boltzmann Transport Equation [0.0]
  Physics-Enhanced Deep Surrogate (PEDS)<n>Network learns geometry-dependent corrections and a mixing coefficient that interpolates between macroscopic and nano-scale behavior.<n>PEDS reduces training-data requirements by up to 70% compared with purely data-driven baselines.
  arXiv  Detail & Related papers  (2025-11-25T16:25:24Z)
• Sparse Model Inversion: Efficient Inversion of Vision Transformers for Data-Free Applications [99.72917069918485]
  We propose a novel sparse model inversion strategy to speed up existing dense inversion methods.<n>Specifically, we invert semantic foregrounds while stopping the inversion of noisy backgrounds and potential spurious correlations.
  arXiv  Detail & Related papers  (2025-10-31T05:14:36Z)
• VIKING: Deep variational inference with stochastic projections [48.946143517489496]
  Variational mean field approximations tend to struggle with contemporary overparametrized deep neural networks.<n>We propose a simple variational family that considers two independent linear subspaces of the parameter space.<n>This allows us to build a fully-correlated approximate posterior reflecting the overparametrization.
  arXiv  Detail & Related papers  (2025-10-27T15:38:35Z)
• Learning from Scratch: Structurally-masked Transformer for Next Generation Lib-free Simulation [5.564681128355971]
  This paper proposes a neural framework for power and timing prediction of multi-stage data path.<n>To the best of our knowledge, this is the first language-based, netlist-aware neural network designed explicitly for standard cells.
  arXiv  Detail & Related papers  (2025-07-23T10:46:25Z)
• Ambient Noise Full Waveform Inversion with Neural Operators [11.44207799108199]
  Recent studies have shown that a new class of machine learning models, called neural operators, can solve the elastodynamic wave equation orders of magnitude faster than conventional methods.<n>In this study, we demonstrate the first application of neural operators for full waveform inversion on a real seismic dataset.
  arXiv  Detail & Related papers  (2025-03-19T09:10:43Z)
• Robust Representation Consistency Model via Contrastive Denoising [83.47584074390842]
  randomized smoothing provides theoretical guarantees for certifying robustness against adversarial perturbations.<n> diffusion models have been successfully employed for randomized smoothing to purify noise-perturbed samples.<n>We reformulate the generative modeling task along the diffusion trajectories in pixel space as a discriminative task in the latent space.
  arXiv  Detail & Related papers  (2025-01-22T18:52:06Z)
• Unified, Verifiable Neural Simulators for Electromagnetic Wave Inverse Problems [1.6795461001108096]
  We show a single, unified model capable of addressing scattering simulations with thousands of DoFs, of any wavelength, any illumination wavefront, and freeform materials. Our findings demonstrate a path to universal, verifiably accurate neural surrogates for existing scattering simulators.
  arXiv  Detail & Related papers  (2024-03-31T03:23:29Z)
• Transform Once: Efficient Operator Learning in Frequency Domain [69.74509540521397]
  We study deep neural networks designed to harness the structure in frequency domain for efficient learning of long-range correlations in space or time. This work introduces a blueprint for frequency domain learning through a single transform: transform once (T1)
  arXiv  Detail & Related papers  (2022-11-26T01:56:05Z)
• Autoencoder-driven Spiral Representation Learning for Gravitational Wave Surrogate Modelling [47.081318079190595]
  We investigate the existence of underlying structures in the empirical coefficients using autoencoders. We design a spiral module with learnable parameters, that is used as the first layer in a neural network, which learns to map the input space to the coefficients. The spiral module is evaluated on multiple neural network architectures and consistently achieves better speed-accuracy trade-off than baseline models.
  arXiv  Detail & Related papers  (2021-07-09T09:03:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.