高能暂现源 · 每日文献综述

HIGH-ENERGY TRANSIENTS · DAILY LITERATURE REVIEW
2024 年 10 月 21 日 星期一 · 数据截至 arXiv / ADS 最新收录日

本页经 1 次补录 ·

I.

今日头条

无
No Breaking · 无突发
今日无通过复核的重大进展

当日 3 篇核心与相关文献均为常规推进,核心 1 篇已按优先级列于下方。

II.

核心文献

1 篇
0110-14 补录
优先级 88 · 多信使触发与联合 · 引力波电磁对应体 · 触发/时域分析/数据产品与方法

Scalable matched-filtering pipeline for gravitational-wave searches of compact binary mergers

面向致密双星合并引力波搜索的可扩展匹配滤波管线

✉ Yun-Jing Huang (Department of Physics, The Pennsylvania State University), Chad Hanna (Department of Physics, The Pennsylvania State University), Becca Ewing (Department of Physics, The Pennsylvania State University) et al.

原文摘要Abstract

As gravitational-wave observations expand in scope and detection rate, the data analysis infrastructure must be modernized to accommodate rising computational demands and ensure sustainability. We present a scalable gravitational-wave search pipeline that modernizes the GstLAL pipeline by adapting the core filtering engine to the PyTorch framework, enabling flexible execution on both central processing units (CPUs) and graphics processing units (GPUs). Offline search results on the same 8.8 day stretch of public gravitational-wave data indicate that the GstLAL and the PyTorch adaptation demonstrate comparable search performance, even with float16 precision. Lastly, computational benchmarking results show that the PyTorch adaptation executed on an A100 GPU achieves speedups of up to 169 times in the GPU float16 configuration and 123 times in the GPU float32 configuration, compared to GstLAL's performance on a single CPU core.

展开 ▾
AI 综述 AI-generated · 以原文为准
亮点

首次演示用PyTorch实现LLOID算法,支持多SVD模板组并行滤波与浮点16精度,在A100 GPU上相比单核CPU最高加速169倍,且搜索性能与原始管线一致。

脉络与展望

匹配滤波是引力波致密双星搜索的核心方法,当前GstLAL等管线采用时域低延迟在线探测算法(LLOID)Cannon+ 2012,但传统的纯CPU实现面对数据量增长面临算力瓶颈。近年来,基于GPU和机器学习的搜索方案显示出扩展潜力Chu+ 2022。本研究将GstLAL的滤波引擎和触发产生器用PyTorch重写,利用GPU并行与半精度运算,在8.8天O3数据上验证了与GstLAL一致的检测效率和注入恢复能力。未来计划现代化管线剩余部分,并借助PyTorch框架融入机器学习技术,以进一步拓展参数空间和提升灵敏度。

预印本 2024-10-21 · 刊出 2025-10-06 · 收录 2026-10-14

边缘相关 2 篇 · 低相关性展开 +
优先级 15 · 引力波 · LISA · 大质量黑洞双星 · 贝叶斯推断

Adapting a novel framework for rapid inference of massive black hole binaries for LISA

Jan, Aasim, O'Shaughnessy, Richard, Shoemaker, Deirdre, et al.

原文摘要Abstract

The Laser Interferometer Space Antenna (LISA) is designed to detect a variety of gravitational-wave events, including mergers of massive black hole binaries, stellar-mass black hole inspirals, and extreme mass-ratio inspirals. LISA's capability to observe signals with high signal-to-noise ratios raises concerns about waveform accuracy. Additionally, its ability to observe long-duration signals will raise the computational cost of Bayesian inference, making it challenging to use costly and novel models with standard stochastic sampling methods without incorporating likelihood and waveform acceleration techniques. In this work, we present our attempt to tackle these issues. We adapt RIFT for LISA to take advantage of its embarrassingly parallel architecture, enabling efficient analysis of large datasets with costly gravitational wave models without relying on likelihood or waveform acceleration. We demonstrate that our open-source code can accurately infer parameters of massive black hole binary signals by carrying out a zero-noise injection recovery using the numerical relativity surrogate model NRHybSur3dq8. By utilizing all available <inline-formula><mml:math><mml:mi>m</mml:mi><mml:mo>!=</mml:mo><mml:mn>0</mml:mn></mml:math></inline-formula> modes in the inference, we study the impact of higher modes on LISA data analysis. We study the impact of multiple massive black hole binary signals in a dataset on the inference of a single signal, showing that the selected source's inference remains largely unaffected. Furthermore, we analyze the LDC-1A and blind LDC-2A datasets from the Radler and Sangria challenge of the LISA data challenges. When eschewing specialized hardware, we find NRHybSur3dq8 injection-recovery takes approximately 20 h to complete, while the analysis of Sangria and Radler datasets takes about 10 h to complete.

展开 ▾

为 LISA 大质量黑洞双星快速推断适配新框架 · 本文将 RIFT 框架适配到 LISA 大质量黑洞双星参数推断,利用并行架构实现无需似然或波形加速的高效分析,并用 NRHybSur3dq8 及 LDC 挑战数据集验证其性能。

预印本 2024-10-21 · 刊出 2025-03-31 · 收录 2026-10-12

优先级 10 · 多信使触发与联合

Approximate helical symmetry in compact binaries

Khairnar, Aniket, Stein, Leo C., Boyle, Michael

原文摘要Abstract

The inspiral of a circular, nonprecessing binary exhibits an approximate helical symmetry. The effects of eccentricity, precession, and radiation reaction break the exact symmetry. We estimate the failure of this symmetry using the flux of the Bondi--van der Burg--Metzner--Sachs (BMS) charge corresponding to helical symmetry carried away by gravitational waves. We analytically compute the helical flux for binaries moving on eccentric orbits and quasicircular orbits without precession using post-Newtonian theory. The helical flux is nonvanishing at the 0PN order for eccentric orbits as expected. We analytically predict the helical flux to be at a relative 5PN order for quasicircular nonprecessing binaries. This prediction is compared with 113 quasicircular nonprecessing numerical relativity waveforms from the SXS catalog. We find good agreement between analytical and numerical results for quasicircular nonprecessing binaries establishing that helical symmetry starts to break at 5PN for these binaries.

展开 ▾

致密双星中的近似螺旋对称性 · 该论文用后牛顿理论计算偏心及准圆无进动双星的BMS螺旋荷通量,并与113个SXS数值相对论波形对比,表明准圆无进动双星的螺旋对称性从5PN阶开始破缺,而偏心轨道的螺旋通量在0PN阶已非零。

预印本 2024-10-21 · 刊出 2025-01-28 · 收录 2026-10-12