Machine learning to assess the astrophysical origin of gravitational waves triggers
机器学习评估引力波触发信号的天体物理起源
In this work, we explore a possible application of a machine learning classifier for candidate events in a template-based search for gravitational-wave signals from various compact system sources. We analyze data from the O3a and O3b data acquisition campaign, during which the sensitivity of ground-based detectors is limited by real non-Gaussian noise transient. The state-of-the-art searches for such signals typically rely on the signal-to-noise ratio (SNR) and a chi-square test to assess the consistency of the signal with an inspiral template. In addition, a combination of these and other statistical properties are used to build ``reweighted SNR'' statistics. We evaluate a Random Forest classifier on a set of double-coincidence events identified using the multiband template analysis pipeline. The new classifier achieves a modest but consistent increase in event detection at low false positive rates relative to the standard search. Using the output statistics from the Random Forest classifier, we compute the probability of astrophysical origin for each event, denoted as <inline-formula><mml:math><mml:msub><mml:mi>p</mml:mi><mml:mtext>astro</mml:mtext></mml:msub></mml:math></inline-formula>. This is then evaluated for the events listed in existing catalogs, with results consistent with those from the standard search. Finally, we search for new possible candidates using these new statistics, with <inline-formula><mml:math><mml:msub><mml:mi>p</mml:mi><mml:mtext>astro</mml:mtext></mml:msub><mml:mo>></mml:mo><mml:mn>0.5</mml:mn></mml:math></inline-formula>, obtaining a new subthreshold candidate (inverse false alarm rate <inline-formula><mml:math><mml:mrow><mml:mo>=</mml:mo><mml:mn>0.05</mml:mn></mml:mrow></mml:math></inline-formula>) event at <inline-formula><mml:math><mml:mi>g</mml:mi><mml:mi>p</mml:mi><mml:mi>s</mml:mi><mml:mo>:</mml:mo><mml:mn>1240423628</mml:mn></mml:math></inline-formula>.
展开 ▾首次将机器学习分类应用于 MBTA 的双符合触发(而非单探测器触发),直接输出 p_s 并计算 p_astro,其效能与标准 ρ_RS_HL 统计量一致,且发现一个 IFAR≈0.05 年的新候选;同时揭示了特征(如 ER)可能对特定事件(GW190924_021846)造成显著偏差。
此前,Kapadia+ 2017 和 Kim+ 2020 等工作展示了机器学习可提升单探测器触发信号的分类性能;本文将方法推广至 MBTA 管道在 O3 运行期间的双符合触发,采用随机森林对多类特征进行整合,输出信号概率 p_s 并基于 KDE 构造天体物理概率 p_astro。结果显示,该方法在低误报率下检测效率与标准排序统计量可比,且对已编录事件的一致性良好,表明机器学习可以作为现有搜索的有效补充。未来可进一步探索无监督学习(如自编码器)用于去噪,或直接利用分类器估算虚假警报率,从而增强实时数据分析的稳健性和自动化水平。
预印本 2025-09-16 · 刊出 2026-04-22 · 收录 2026-10-12