PulseAugur
EN
LIVE 13:38:17

PINNs recover conductivity features with 3-12% error using wavelets and FFE · 2 sources tracked

By PulseAugur Editorial · [2 sources] ·

Researchers have developed a new framework using physics-informed neural networks (PINNs) to reconstruct conductivity features from limited boundary data in the Calderón inverse problem. The method incorporates randomized wavelet functions and Fourier-feature encoding to better represent sharp variations in conductivity. Evaluations using synthetic data show the framework can recover dominant conductivity structures with relative errors between 3% and 12%, with Fourier-feature encoding proving particularly effective for localized sharp features like inclusions and interfaces. AI

IMPACT This research advances the application of neural networks in solving complex inverse problems, potentially improving subsurface imaging and material characterization.

RANK_REASON Academic paper detailing a new method for solving an inverse problem using neural networks.

Read on arXiv cs.LG →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

PINNs recover conductivity features with 3-12% error using wavelets and FFE · 2 sources tracked

COVERAGE [2]

  1. arXiv cs.LG TIER_1 English(EN) · Ali AlHadi Kalout, Pablo Tejerina-P\'erez, Konstantin Karchev, Pedro Taranc\'on-\'Alvarez, Leonid Sarieddine, Raul Jimenez, Max Engelstein, Guy David ·

    Recovering Sharp Conductivity Features in the Finite-Data Calder\'on Problem with Physics-Informed Neural Networks

    arXiv:2606.28158v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have recently emerged as a promising framework for addressing the Calder\'on inverse problem from limited boundary data. In this work, we revisit neural Calder\'on inversion by introducing mu…

  2. arXiv cs.LG TIER_1 English(EN) · Guy David ·

    Recovering Sharp Conductivity Features in the Finite-Data Calderón Problem with Physics-Informed Neural Networks

    Physics-informed neural networks (PINNs) have recently emerged as a promising framework for addressing the Calderón inverse problem from limited boundary data. In this work, we revisit neural Calderón inversion by introducing multiscale boundary excitations based on randomized wa…

RELATED ENTITIES

RELATED TOPICS