arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.15948math.OCcs.NAmath.NA
收藏

基于临界性度量的无限维优化误差估计

Criticality measure-based error estimates for infinite dimensional optimization

Danlin Li, Johannes Milz

AI总结:

针对无限维 Hilbert 空间优化问题,提出基于临界性度量的近似框架,证明了阶数最优的误差估计,不依赖强凸性或二阶条件,并在 PDE 约束优化中验证了理论收敛速率。

摘要:

受微分方程优化的启发,我们考虑以 Hilbert 空间为决策空间的优化问题。由于其无限维特性,数值求解需要有限维近似和离散化。我们开发了一个近似框架,并展示了基于临界性度量的误差估计。我们考虑的临界性度量受到优化方法(如半光滑 Newton 法和(条件)梯度法)中使用的度量的启发。此外,我们证明了我们的误差估计是阶数最优的。我们的研究结果扩充了现有的基于距离的误差估计,但不依赖强凸性或二阶充分最优性条件。此外,我们的误差估计可用于代码验证与确认。我们在线性、半线性和双线性 PDE 约束优化上说明了我们的理论收敛速率。

英文摘要:

Motivated by optimization with differential equations, we consider optimization problems with Hilbert spaces as decision spaces. As a consequence of their infinite dimensionality, the numerical solution necessitates finite dimensional approximations and discretizations. We develop an approximation framework and demonstrate criticality measure-based error estimates. We consider criticality measures inspired by those used within optimization methods, such as semismooth Newton and (conditional) gradient methods. Furthermore, we show that our error estimates are order-optimal. Our findings augment existing distance-based error estimates, but do not rely on strong convexity or second-order sufficient optimality conditions. Moreover, our error estimates can be used for code verification and validation. We illustrate our theoretical convergence rates on linear, semilinear, and bilinear PDE-constrained optimization.

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。

↑