arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1405.2673stat.COstat.AP
收藏

粒子MCMC用于贝叶斯微波控制

Particle MCMC for Bayesian Microwave Control

P. Minvielle, A. Todeschini, F. Caron, P. Del Moral

摘要:

我们考虑从全球电磁散射测量的局部无线电特性估计的问题。这个具有挑战性的不适定高维反问题可以探索密集的计算并行麦克斯韦求解器上的petaflopic超级计算机。然后,它示出了如何贝叶斯推断可以执行与粒子边际大都市黑斯廷斯(PMMH)的方法,其中包括一个饶-布莱克韦利斯序贯蒙特卡罗算法与交互卡尔曼滤波器。材料特性,包括多个组件的“德拜弛豫”/“洛伦兹共振”材料模型,估计,它说明了合成数据。最终,我们提出了不同的方法来处理高维问题,从并行化到最初引入的高效的顺序数据同化技术,广泛应用于天气预报,海洋学,地球物理学等。

英文摘要:

We consider the problem of local radioelectric property estimation from global electromagnetic scattering measurements. This challenging ill-posed high dimensional inverse problem can be explored by intensive computations of a parallel Maxwell solver on a petaflopic supercomputer. Then, it is shown how Bayesian inference can be perfomed with a Particle Marginal Metropolis-Hastings (PMMH) approach, which includes a Rao-Blackwellised Sequential Monte Carlo algorithm with interacting Kalman filters. Material properties, including a multiple components "Debye relaxation"/"Lorenzian resonant" material model, are estimated; it is illustrated on synthetic data. Eventually, we propose different ways to deal with higher dimensional problems, from parallelization to the original introduction of efficient sequential data assimilation techniques, widely used in weather forecasting, oceanography, geophysics, etc.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑