NRHJSur3dq8：非偏心对齐自旋双黑洞波形的一个谱数值相对论代理族
NRHJSur3dq8: a spectral numerical-relativity surrogate family for non-eccentric aligned-spin binary-black-hole waveforms
- Institute for Gravitation and the Cosmos, The Pennsylvania State University（宇宙与引力研究所，宾夕法尼亚州立大学）
- Department of Physics, The Pennsylvania State University（物理系，宾夕法尼亚州立大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出NRHJSur3dq8代理族，基于作用-角参数化，用高斯过程回归和切比雪夫单元高效生成非偏心对齐自旋双黑洞波形，精度高且支持近解析导数和不确定性估计。
AI中文摘要：
我们提出NRHJSur3dq8，一个受作用-角参数化启发的非偏心对齐自旋双黑洞波形数值相对论（NR）代理族。该模型同时提供动力学量：哈密顿量值、方位作用$J_\phi(\Phi)$、轨道时钟$\tau(\Phi)$作为其相位的函数，以及辐射量：波形、其近解析导数以及与之相关的建模误差。代理变量被分解为在辐射反应时标上缓慢变化的部分，表示在共享的$hp$自适应切比雪夫单元上，并通过高斯过程模型在固有参数上进行回归。在绝热描述失效处附加一组时间参数化的并合-铃宕单元。对中心NRHJSur3dq8_AA变体对NR的留一验证在完整旋进-并合-铃宕跨度上达到$(2,2)$非加权失配$1.3^{+32}_{-1.2}\times10^{-6}$和全模式失配$2.3^{+59}_{-2.2}\times10^{-6}$。对每次模拟的$32$个方向进行天空平均和信噪比加权，发布模型对用于训练NR模拟的全IMR样本内失配在总质量$40$-$120\\,M_\odot$范围内为$8.9\times10^{-7}$至$2.8\times10^{-5}$。波形使用后牛顿波形进行混合。除精度外，NRHJSur3dq8提供了NR代理中不常见的两种能力：应变对物理参数的近解析导数，以及模型的校准高斯过程预测不确定性。完整的非混合旋进-并合-铃宕波形在单个CPU调用上评估需$12$毫秒。批量评估在批量大小$4096$下摊销至每波形$0.39$毫秒，在起始频率$20$ Hz下测量。
英文摘要：
We present NRHJSur3dq8, a family of numerical-relativity (NR) surrogates for non-eccentric aligned-spin binary-black-hole waveforms inspired by an action-angle parameterization. The model provides both the dynamics: the Hamiltonian values, azimuthal action $J_ϕ(Φ)$, the orbital clock $τ(Φ)$ as functions of its phasing, and the radiative pieces: the waveforms, their near-analytic derivatives, and modeling errors associated with each. The surrogated variables are conditioned into pieces that vary slowly on the radiation-reaction timescale, represented on shared $hp$-adaptive Chebyshev elements, and regressed across the intrinsic parameters with a Gaussian-process model. A time-parameterized merger-ringdown set of elements is attached where the adiabatic description breaks down. Leave-one-out validation of the central NRHJSur3dq8_AA flavor against NR reaches a $(2,2)$ unweighted mismatch of $1.3^{+32}_{-1.2}\times10^{-6}$ and an all-mode mismatch of $2.3^{+59}_{-2.2}\times10^{-6}$ over the full inspiral-merger-ringdown span. Sky-averaged and SNR-weighted over $32$ orientations per simulation, the released model's full-IMR in-sample mismatch against the NR simulations used for training is $8.9\times10^{-7}$ to $2.8\times10^{-5}$ across total masses $40$-$120\,M_\odot$. Waveforms are hybridized using post-Newtonian waveforms. Beyond accuracy, NRHJSur3dq8 provides two capabilities uncommon in NR surrogates: nearly analytic derivatives of the strain with respect to the physical parameters, and a calibrated Gaussian-process predictive uncertainty of the model. A full non-hybridized inspiral-merger-ringdown waveform evaluates in $12$ ms on a single CPU call. Batched evaluation amortizes this to $0.39$ ms per waveform at a batch of $4096$, measured at a starting frequency of $20$ Hz.
