Stream SGD的收敛性与推断,及其在排队系统和库存控制中的应用
Convergence and Inference of Stream SGD, with Applications to Queueing Systems and Inventory Control
AI总结:
本文构建stream SGD的系统分析框架,利用Poisson方程处理Markov采样偏差,证明O(1/T)收敛与O(log T)遗憾,并给出在线推断及排队、库存应用。
摘要:
Stream随机梯度下降(SGD)是求解运筹学(OR)中在线优化问题的一种简单高效的方法,在这类问题中,数据由依赖参数的Markov链生成。不同于在迭代过程中需要不断增大批量大小的传统方法,stream SGD在每次迭代中仅使用单个样本,从而显著提升样本效率。本文建立了一个用于分析stream SGD的系统性框架,利用Poisson方程解来处理梯度偏差和统计相关性。我们证明了最优的O(1/T)收敛速率以及当前最先进的O(log T)遗憾,同时引入了一种用于不确定性量化的在线推断方法,并通过一个新的泛函中心极限定理为其提供支撑。我们提出了一种新的Wasserstein型散度来描述该框架的条件,这使得相关假设能够通过为底层OR模型量身定制的耦合技术得到直接验证。我们考虑了其在排队系统和库存管理中的应用,展示了该方法的实用性和广泛相关性,并为stream SGD在OR领域中的有效性提供了新的见解。
英文摘要:
Stream stochastic gradient descent (SGD) is a simple and efficient method for solving online optimization problems in operations research (OR), where data is generated by parameter-dependent Markov chains. Unlike traditional approaches which require increasing batch sizes during iterations, stream SGD uses a single sample per iteration, significantly improving sample efficiency. This paper establishes a systematic framework for analyzing stream SGD, leveraging the Poisson equation solution to address gradient bias and statistical dependence. We prove optimal O(1/T) convergence rates and the state-of-the-art O(log T) regret, while also introducing an online inference method for uncertainty quantification and supporting it by a novel functional central limit theorem. We propose a novel Wasserstein-type divergence to describe the framework's conditions, which makes the assumptions in question directly verified via coupling techniques tailored to underlying OR models. We consider applications in queueing systems and inventory management, demonstrating the practicality and broad relevance, as well as providing new insights into the effectiveness of stream SGD in OR fields.
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。