发表机构
INFN Sezione di Ferrara（意大利国家核物理研究所费拉拉分部）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出一种基于确定性扩散型生成模型（U-Net 实现 $\alpha$-(去)混合）的机器学习方法，用于对连续引力波候选信号进行去噪，在信噪比 $\rho \gtrsim 4$ 时能有效恢复信号模式，可作为搜索流水线中的一致性/否决工具。
AI 中文摘要
我们提出一种确定性的机器学习方法，用于对来自旋转中子星连续引力波全天搜索的候选信号进行去噪。利用时域 $F$-统计量搜索流水线，我们通过迭代 $\alpha$-(去)混合（一种以 U-Net 实现的确定性扩散型生成模型，非常适合 $F$-统计量输出的非高斯、相关噪声）对所得的 $F(f,\dot{f})$ 候选模式进行后处理。两个模型分别基于模拟的 6 天和 12 天时域段数据训练，这些数据包含与 LVK 硬件注入相似的软件注入信号，并通过将去混合图像与预期信号模式库进行比较（使用结构相似性指数度量）进行测试。总体而言，该方法能够恢复信噪比 $\rho \gtrsim 4$（略低于典型全天搜索检测阈值）下正确的、依赖于天空位置的模式，恢复效果强烈依赖于模式形态，这表明基于确定性扩散的去噪方法可以作为半相干符合阶段之前的一致性/否决工具。我们还讨论了该方法的局限性和可能的改进方向。
英文摘要
We present a deterministic machine-learning method for denoising candidate signals from all-sky searches for continuous gravitational waves from rotating neutron stars. Using the time-domain $F$-statistic search pipeline, we post-process the resulting $F(f,\dot{f})$ candidate patterns with iterative $α$-(de)blending, a deterministic diffusion-type generative model implemented as a U-Net, well suited to the non-Gaussian, correlated noise of the $F$-statistic output. Two models, trained on data based on simulated 6- and 12-day time-domain segments with software-injected signals similar to the LVK hardware injections, are tested by comparing deblended images to the library of expected signal patterns via the Structural Similarity Index Measure. In general the method recovers the correct sky-position-dependent pattern for signal-to-noise ratios $ρ\gtrsim 4$, moderately below typical all-sky detection thresholds, with recovery depending strongly on pattern morphology, demonstrating that deterministic diffusion-based denoising may serve as a consistency/veto tool ahead of the semi-coherent coincidence stage. We also discuss limitations and possible improvements of the method.
Comments23 pages, 12 figures. Submitted to Classical and Quantum Gravity
