Id: arxiv/2607.03772 Url: https://arxiv.org/abs/2607.03772 Tags: math.NA Score: 0 Title: Level-set physics-informed neural networks for domain inverse problems of gravimetry Domain: math.NA Source: arxiv Authors: Jingnan Yao, Wenbin Li, Jianliang Qian Reviews: 0 Subject: math.NA Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-07-04 Page: https://delphi.science/r/arxiv/2607.03772 We propose level-set physics-informed neural networks (PINNs) for domain inverse problems of gravimetry. The domain inverse problem establishes a correctness class for ill-posed inverse gravimetry, which we solve within the PINNs framework. Directly representing the domain inverse problem via neural networks is problematic due to the discontinuous nature of interfaces. We consider a level-set formulation where the neural network represents a continuous level-set function, and its zero level-set depicts sharp interfaces. To overcome the challenges of exploding and vanishing gradients caused…