声波全波形反演中神经先验的结构依赖失效模式
Structure-dependent failure modes of neural priors in acoustic full-waveform inversion
另 1 家 · 查看机构详情
- Institute of Geophysics, China Earthquake Administration（中国地震局地球物理研究所）
- Laboratory of Seismology and Physics of Earth’s Interior, School of Earth and Space Sciences, University of Science and Technology of China（中国科学技术大学地球和空间科学学院地球深部探测实验室）
- Institute of Advanced Technology, University of Science and Technology of China（中国科学技术大学先进技术研究院）
- University of Chinese Academy of Sciences（中国科学院大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
该研究比较了三种神经先验在声波全波形反演中的表现，发现频率延拓并非普遍有效，且模型恢复与波形预测无必然关联，主张联合报告多种指标以审计神经FWI。
中文摘要 AI 辅助
频率延拓并不能改善每一种神经表示，且更好的速度恢复并不必然意味着对未见波形的更好预测。我们在二维声波全波形反演中比较了带全变差正则化的速度网格、自适应正弦坐标网络（SIREN）以及带全网格残差的冻结图像生成器。所有方法使用相同的冻结SWEEP离散化、采集、观测、背景和1500次炮-模型梯度评估的预算。三个暴露的合成目标代表倾斜层状、弯曲层状以及来自断层标注族的高对比度弯曲界面。在2至10赫兹的延拓下，生成-残差模型的速度均方根误差分别为218.7、108.0和341.9米/秒，而网格加TV方法分别为232.5、100.3和412.9米/秒。因此，后者在弯曲层状目标上获胜，而前者在其他两个目标上将模型均方根误差分别提高了5.9%和17.2%。然而，网格加TV在所有三个目标上对保留的10赫兹炮的预测更好。SIREN在每个单元上均差于公共背景；延拓使两个目标变差但改善了一个目标。一项五权重TV敏感性研究保持了曲线断层模型误差的排名。一个等预算的变分扩展在恢复和区间覆盖上均失败。这些结果确立了条件性排名，而非普遍优越性或孤立的谱偏置机制。全网格残差也阻止了将收益单独归因于生成器。我们主张在审计神经FWI时联合报告模型恢复、未见炮预测、优化成本和负面结果。
英文摘要
Frequency continuation does not improve every neural representation, and better velocity recovery need not imply better prediction of unseen waveforms. We compare a velocity grid with total-variation regularization, an adapted sinusoidal coordinate network (SIREN), and a frozen image generator with a full-grid residual in two-dimensional acoustic full-waveform inversion. All methods use the same frozen SWEEP discretization, acquisition, observations, background, and budget of 1,500 shot--model gradient evaluations. Three exposed synthetic targets represent inclined layering, curved layering, and a high-contrast curved interface drawn from a fault-labelled family. With 2-to-10 Hz continuation, the generative-residual model attains velocity RMSEs of 218.7, 108.0, and 341.9 m/s, compared with 232.5, 100.3, and 412.9 m/s for Grid+TV. Thus the latter wins on curved layering, while the former improves model RMSE by 5.9% and 17.2% on the other targets. Grid+TV nevertheless predicts held-out 10 Hz shots better on all three. SIREN is worse than the common background in every cell; continuation worsens two targets but improves one. A five-weight TV sensitivity study preserves the CurveFault model-error ranking. An equal-budget variational extension fails both recovery and interval coverage. These results establish conditional rankings, not universal superiority or an isolated spectral-bias mechanism. Full-grid residuals also prevent attributing gains uniquely to the generator. We advocate joint reporting of model recovery, unseen-shot prediction, optimization cost, and negative outcomes when auditing neural FWI.
