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~~
arXiv 2608.22201stat.MEcs.LG

用于扫描统计的高效回归模型

Efficient Regression Models for Scan Statistics

  • University of Utah(犹他大学)
  • CosmicAI(宇宙人工智能公司)
  • National Radio Astronomy Observatory(国家射电天文台)

机构由 AI 辅助整理,请以论文原文为准。

Gazi Abdur Rakib, Tristan Ashton, Ryan A. Loomis, Brian S. Mason, Eric J. Murphy, Ci Xue, Jeff M. Phillips

AI总结:

该研究针对扫描统计提出新型高效回归模型,将原O(n⁴)复杂度优化为线性时间,可有效检测合成异常与干涉天文学的真实平台化问题。

AI中文摘要:

我们提出了一类针对实值信号扫描统计的新型回归模型,这类模型可更好地拟合非平稳信号,以对比扫描统计识别出的区间异常。我们的模型可表示广义似然比统计量。对于长度为n的信号,这类方法原本需要O(n⁴)的计算量,我们通过算法改进得到了线性时间算法(假设最大区间宽度受限)。我们的方法,尤其是基于Nadaraya-Watson核回归的方法,在检测人工植入的异常以及识别干涉天文学中真实的“平台化”问题时表现尤为出色。

英文摘要:

We introduce a new class of regression models for scan statistics on real-valued signals. These allow for improved fitting of non-stationary signals to contrast with the interval anomalies identified by the scan statistics. Our models can represent generalized likelihood ratio statistics. While these methods naively require $O(n^4)$ for a length $n$ signal, we provide algorithmic improvements which lead to linear time algorithms (with assumptions on max interval width). Our methods, especially ones based on Nadaraya-Watson kernel regression, are demonstrated as especially effective in detecting both synthetically planted anomalies, and for identifying a real ``platforming'' issue in interferometric astronomy.

补充信息

↑