面向未归一化模型的鲁棒最快变化检测
Robust Quickest Change Detection for Unnormalized Models
AI总结:
针对在线数据流中变化后分布未知且模型未归一化的问题,本文提出基于分数的鲁棒 RSCUSUM 算法,用 Fisher 散度替代 KL 散度并引入最不利分布,仿真验证了其性能与理论分析。
摘要:
检测在线数据流底层分布中突发且持续的变化,是许多应用中的重要问题。本文提出一种新的基于分数的鲁棒算法 RSCUSUM,该算法可应用于未归一化模型,并解决变化后分布未知的问题。RSCUSUM 将变化前与变化后分布之间的 Kullback-Leibler 散度替换为 Fisher 散度,以提高未归一化统计模型中的计算效率,并引入用于鲁棒变化检测的“最不利”分布概念。该算法及其理论分析通过仿真研究得到验证。
英文摘要:
Detecting an abrupt and persistent change in the underlying distribution of online data streams is an important problem in many applications. This paper proposes a new robust score-based algorithm called RSCUSUM, which can be applied to unnormalized models and addresses the issue of unknown post-change distributions. RSCUSUM replaces the Kullback-Leibler divergence with the Fisher divergence between pre- and post-change distributions for computational efficiency in unnormalized statistical models and introduces a notion of the ``least favorable'' distribution for robust change detection. The algorithm and its theoretical analysis are demonstrated through simulation studies.
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。