arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2304.04550cs.DSmath.OC
收藏

使用梯度预言机的内点法

Interior Point Methods with a Gradient Oracle

Adrian Vladu

AI总结:

本文提出一种仅需一阶梯度预言机的拟牛顿内点法,通过扩展Dunagan-Harvey技术维护预条件子,在凸集优化和半定规划上实现了优于现有算法的运行时间。

摘要:

我们提出了一种基于拟牛顿迭代的内点法,该方法仅需对强自协调障碍函数进行一阶访问。为此,我们扩展了Dunagan-Harvey [STOC '07]的技术来维护一个预条件子,同时仅使用一阶信息。我们通过该预条件子相对于未知Hessian矩阵的相对离心率来衡量其质量,并将这些技术推广到具有缓慢变化Hessian矩阵的凸函数。我们将此与内点法相结合,证明在给定对凸集 $K$ 的适当障碍函数的一阶访问权限时,可以在 $\widetilde{O}\left(\left(\mathcal{T}+n^{2}\right)\sqrt{nν}\log\left(1/\varepsilon\right)\right)$ 时间内求解 $K$ 上的良态线性优化问题至 $\varepsilon$ 精度,其中 $ν$ 是障碍函数的自协调参数,$\mathcal{T}$ 是进行梯度查询所需的时间。由此我们证明:\n $\bullet$ $n$ 维凸集上的线性优化可在 $\widetilde{O}\left(\left(\mathcal{T}n+n^{3}\right)\log\left(1/\varepsilon\right)\right)$ 时间内求解。当用适当障碍函数的一阶预言机替代分离预言机时,这与最先进的割平面法所实现的运行时间相当。\n $\bullet$ 我们可以在 $\widetilde{O}\left(mn^{4}+m^{1.25}n^{3.5}\log\left(1/\varepsilon\right)\right)$ 时间内求解涉及 $\mathbb{R}^{n\times n}$ 中 $m\geq n$ 个矩阵的半定规划,在 $m=Ω\left(n^{\frac{3.5}{ω-1.25}}\right)$ 的情况下优于现有最先进算法。\n 在此过程中,我们开发了一系列工具,利用矩阵分析和Schur凸性技术来控制势函数的演化。

英文摘要:

We provide an interior point method based on quasi-Newton iterations, which only requires first-order access to a strongly self-concordant barrier function. To achieve this, we extend the techniques of Dunagan-Harvey [STOC '07] to maintain a preconditioner, while using only first-order information. We measure the quality of this preconditioner in terms of its relative excentricity to the unknown Hessian matrix, and we generalize these techniques to convex functions with a slowly-changing Hessian. We combine this with an interior point method to show that, given first-order access to an appropriate barrier function for a convex set $K$, we can solve well-conditioned linear optimization problems over $K$ to $\varepsilon$ precision in time $\widetilde{O}\left(\left(\mathcal{T}+n^{2}\right)\sqrt{nν}\log\left(1/\varepsilon\right)\right)$, where $ν$ is the self-concordance parameter of the barrier function, and $\mathcal{T}$ is the time required to make a gradient query. As a consequence we show that: $\bullet$ Linear optimization over $n$-dimensional convex sets can be solved in time $\widetilde{O}\left(\left(\mathcal{T}n+n^{3}\right)\log\left(1/\varepsilon\right)\right)$. This parallels the running time achieved by state of the art algorithms for cutting plane methods, when replacing separation oracles with first-order oracles for an appropriate barrier function. $\bullet$ We can solve semidefinite programs involving $m\geq n$ matrices in $\mathbb{R}^{n\times n}$ in time $\widetilde{O}\left(mn^{4}+m^{1.25}n^{3.5}\log\left(1/\varepsilon\right)\right)$, improving over the state of the art algorithms, in the case where $m=Ω\left(n^{\frac{3.5}{ω-1.25}}\right)$. Along the way we develop a host of tools allowing us to control the evolution of our potential functions, using techniques from matrix analysis and Schur convexity.

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。

↑