.png){kind=icon}
返回
SEISMIC TOMOGRAPHY WITH RANDOM BATCH GRADIENT RECONSTRUCTION
.png){kind=icon}
OA
AI
DOI：10.1137/21M1452342.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Seismic tomography solves high-dimensional optimization problems in the imaging of Earth's subsurface structures. In this paper, we propose using random batch methods to construct the gradient used for iterations in seismic tomography. Specifically, we use the frozen Gaussian approximation to compute seismic wave propagation and construct stochastic gradients by random batch methods. This method inherits the spirit of stochastic gradient descent methods for solving high-dimensional optimization problems. The proposed idea is general in the sense that it does not rely on the use of frozen Gaussian approximation, and one can replace it with any other efficient wave propagation solver, e.g., Gaussian beam methods and spectral element methods. We prove the convergence of the random batch method in the mean-square sense and show the numerical performance of the proposed method by two- and three-dimensional examples of wave-equation-based travel-time inversion and full-waveform inversion, respectively. As a by-product, we also prove the convergence of the accelerated full-waveform inversion using dynamic mini-batches and spectral element methods.
Keyword:
seismic tomography
random batch method
frozen Gaussian approximation
high-dimensional optimization
inverse problems
期刊
IF:
2.6
论文数:
5.1K
被引数:
1.8W
机构
引用论文
A multi-step process to improve texturing and transport current in BSCCO (2223) superconductors一种多步工艺，用于改善BSCCO (2223)超导体的织构和传输电流
.png){kind=icon}
PREAI
分享收藏分享收藏Apparent strip-like behavior of superconducting tape due to jc(B) dependence超导带材由于Jc(B)依赖性表现出明显的条带状行为
PREAI
分享收藏A frozen Gaussian approximation-based multi-level particle swarm optimization for seismic inversion
OAAI
OAAI分享收藏
查看更多内容