arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.00709cs.LGmath.OC
收藏

驯服黎曼随机近似

Tame Riemannian Stochastic Approximation

Johannes Aspman, Vyacheslav Kungurtsev, Reza Roohi Seraji

AI总结:

本文首次研究黎曼流形上的驯服优化问题,证明驯服函数结构能为随机次梯度下降提供理论保证,并通过Retracted SGD算法分析验证了其收敛性与适用性。

摘要:

我们研究了随机近似应用于受黎曼流形定义约束的驯服非可微函数的性质。源于o-minimal拓扑学并在推广至流形时扩展到几何范畴的驯服函数,其目标景观展现出某种结构,该结构为通用随机次梯度下降(sub-gradient descent)提供了期望函数下降和渐近收敛的理论保证。近期工作表明,此类函数能忠实建模深度神经网络训练目标的损失景观,且深度学习包中使用的autograd操作实现了具有正确收敛性质的次梯度下降变体。黎曼优化利用约束集的几何性质执行最小化过程,同时强制优化变量保持在黎曼流形上。本文首次研究了黎曼流形上的驯服优化,突出了该问题丰富的几何结构,并通过简单的Retracted SGD算法的分析与数值报告,证实了规范“SGD”对这类问题的适用性。

英文摘要:

We study the properties of stochastic approximation applied to a tame nondifferentiable function subject to constraints defined by a Riemannian manifold. The objective landscape of tame functions, arising in o-minimal topology extended to a geometric category when generalized to manifolds, exhibits some structure that enables theoretical guarantees of expected function decrease and asymptotic convergence for generic stochastic sub-gradient descent. Recent work has shown that this class of functions faithfully model the loss landscape of deep neural network training objectives, and the autograd operation used in deep learning packages implements a variant of subgradient descent with the correct properties for convergence. Riemannian optimization uses geometric properties of a constraint set to perform a minimization procedure while enforcing adherence to the the optimization variable lying on a Riemannian manifold. This paper presents the first study of tame optimization on Riemannian manifolds, highlighting the rich geometric structure of the problem and confirming the appropriateness of the canonical "SGD" for such a problem with the analysis and numerical reports of a simple Retracted SGD algorithm.

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。

↑