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~~

光滑全局PLI函数是非线性最小二乘,其梯度主导的同类函数亦是如此

Smooth globally PLI functions are nonlinear least-squares, and so are their gradient-dominated cousins

Eduardo D. Sontag

arXiv 2608.08849首次发表:更新:

AI 中文总结

该研究针对BCR的全局PLI不等式假设过强的问题,提出用sgl-PLI替换gl-PLI,保留原结构结论,且该结论适用于连续时间LQR策略优化和逻辑回归问题。

AI 中文摘要

Boumal、Criscitiello和Rebjock(BCR)在[arXiv:2604.07972, 2026]中证明,在可缩完备黎曼流形上定义的满足(全局)Polyak-Lojasiewicz不等式(PLI)的光滑实值函数f,必然具有f = f* + φ²的形式,其中φ为淹没映射,同时还得到了一长串推论。然而,在一些最需要该假设的问题中,这一假设并不成立,其中包括连续时间LQR策略优化和逻辑回归,这也是多篇论文中开发广义PLI不等式层级的动机所在。我们在此观察到,全局PLI不等式比BCR论文证明所需的条件更强。将其替换为一个正定函数,该函数在原点附近由√的倍数下界有界,也就是将gl-PLI替换为“sgl-PLI”,可保留所有结构结论不变。而两种直接的弱化形式均失效,相关基本结论适用于连续时间LQR问题和逻辑回归。

英文摘要

Boumal, Criscitiello and Rebjock (BCR) proved that if $M$ is a contractible, connected and complete Riemannian manifold, then every smooth function $f\colon M\to R$ satisfying the global Polyak--Łojasiewicz inequality (PŁI) is necessarily of the form $f = f^* + \|ϕ\|^2$ with $ϕ$ a submersion. Informally, minimizing such a function amounts to solving a nonlinear least-squares problem in new coordinates. The global PŁI hypothesis fails, however, in many problems of interest, among them continuous-time LQR policy optimization in optimal control and a standard formulation of logistic regression. A hierarchy of weakened PŁ inequalities has been introduced in order to cover such problems, and more generally to study the effect of noise and adversarial perturbations on gradient flows. This note shows that, with minor modifications, the same reduction to a nonlinear least-squares problem holds under a substantially weaker hypothesis, ``semiglobal'' PŁI, which is satisfied in both of the examples just mentioned. That condition asks that $f$ satisfy an estimate $\|\nabla f(x)\| \ge α\bigl(f(x)-f^*\bigr)$ for all $x$, with $α$ merely positive definite and bounded below by a positive multiple of $\sqrt{s}$ for small $s>0$.

论文原文

arXiv 摘要页 · PDF 原文 · HTML 原文

↑