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 2406.14808math.STcs.LGstat.MEstat.MLstat.TH
收藏

关于贝叶斯PINN求解反问题的估计速率

On the estimation rate of Bayesian PINN for inverse problems

Boston University(波士顿大学)

Yi Sun, Debarghya Mukherjee, Yves Atchade

AI总结:

该研究针对贝叶斯物理信息神经网络(PINNs)求解参数线性偏微分方程反问题的性能,推导了后验均方误差的下界及未知线性系数的收敛速率,并通过仿真验证了理论结果。

摘要:

在物理与机器学习领域,使用物理信息神经网络(Physics-informed neural networks, PINNs)求解偏微分方程(partial differential equations, PDEs)及其反问题是一种发展迅速的方法。尽管已有多种PINNs架构在实际应用中表现出色,但我们对其性能的理论理解仍较为有限。本文研究了基于n次独立带噪解测量值的贝叶斯PINN估计器求解PDE的行为,聚焦一类参数线性(含未知系数θ_⋆)的方程。研究表明,当偏微分方程存在β阶可微的经典解(记为u_⋆)时,贝叶斯后验均值的均方误差至少为n^{-2β/(2β+ d)}量级。此外,我们还建立了θ_⋆线性系数的收敛速率,该速率取决于底层微分算子的阶数。最后,我们通过大量仿真验证了理论结果的正确性。

英文摘要:

Solving partial differential equations (PDEs) and their inverse problems using Physics-informed neural networks (PINNs) is a rapidly growing approach in the physics and machine learning community. Although several architectures exist for PINNs that work remarkably in practice, our theoretical understanding of their performances is somewhat limited. In this work, we study the behavior of a Bayesian PINN estimator of the solution of a PDE from $n$ independent noisy measurement of the solution. We focus on a class of equations that are linear in their parameters (with unknown coefficients $θ_\star$). We show that when the partial differential equation admits a classical solution (say $u_\star$), differentiable to order $β$, the mean square error of the Bayesian posterior mean is at least of order $n^{-2β/(2β+ d)}$. Furthermore, we establish a convergence rate of the linear coefficients of $θ_\star$ depending on the order of the underlying differential operator. Last but not least, our theoretical results are validated through extensive simulations.

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。

↑