摊销不透明度边缘化改进褐矮星反演的C/O区间校准
Amortized Opacity Marginalization Improves C/O Interval Calibration for Brown-Dwarf Retrievals
浏览论文内容
中文总结 AI 辅助
该研究针对褐矮星反演中分子线表不一致导致的C/O区间校准误差问题，提出摊销不透明度边缘化方法，通过模拟训练集随机化不透明度，使神经后验估计器学习边缘化后验，提升了C/O区间覆盖率。
中文摘要 AI 辅助
分子线表的不一致程度已可显著改变反演得到的碳氧比（C/O），因此基于单一线表开展的反演会在其误差棒中遗漏一个主导性系统误差。经典边缘化方法是对每个天体在K种不透明度实现下重新运行完整反演。我们将这一成本转移到模拟中：在生成训练集时，我们随机化不透明度（在logν中使用平滑的乘法傅里叶场，加上离散的线表混合，采用由实测线表间差异确定的故意宽泛先验），从而让神经后验估计器（NPE）学习到已完成边缘化的后验。从概念上讲，这是在多个合理不透明度实现下重复常规反演并对所得后验取平均的摊销模拟。在2048个扰动的保留光谱上，标准NPE覆盖名义90% C/O区间的比例为43.3%，而计算量匹配的边缘化网络的该比例为82.4%；三次随机种子控制集成在干净数据上达到93.2%，但在扰动下为61.4%，而边缘化网络为89.4%，这将缺陷归因于不透明度诱导的（TARP联合期望覆盖偏差为0.005，对比0.035）。轴消融分析表明，增益来自连续场（仅离散交换时为45.4%，仅场时为79.3%）。应用105个真实坏像素掩码并采用中值填充后，集成的C/O覆盖率从0.901降至0.479；在相同掩码分布上重新训练并合并三个此类种子后，在完整部署路径下（1024个样本，相同105个掩码），名义C/O为0.905，整体覆盖率为0.891，处于先验宽度的0.21至0.31倍。在13颗晚型T/Y型矮星的105个JWST/G395H光谱产品上，单一边缘化网络划定了Hood等人（2024）T8基准C/O的区间，而控制组未包含该基准，这是一致性检查而非验证，区间宽度为其1.6至2.6倍。
英文摘要
Molecular line lists disagree at a level that measurably shifts retrieved C/O, so retrievals conditioned on a single list omit a dominant systematic from their error bars. Classical marginalization re-runs a full retrieval under K opacity realizations per object. We move that cost into simulation: while generating the training set we randomize opacities (smooth multiplicative Fourier fields in $\logν$ plus a discrete line-list mixture, under a deliberately broad prior informed by measured inter-list differences), so a neural posterior estimator learns an already-marginalized posterior. Conceptually, this is the amortized analogue of repeating a conventional retrieval under many plausible opacity realizations and averaging the resulting posteriors. On perturbed held-out spectra (n=2048) a standard NPE covers nominal 90% C/O intervals 43.3% of the time against 82.4% for a compute-matched marginalized network; three-seed control ensembles reach 93.2% on clean data but 61.4% under perturbation, against 89.4% marginalized, isolating the deficit as opacity-induced (TARP joint expected-coverage deviation 0.005 against 0.035 for a single control network). An axis ablation traces the gain to the continuous field (45.4% for the discrete swaps alone, 79.3% for the field alone). Applying the 105 real bad-pixel masks with median filling then collapsed the ensemble's C/O coverage from 0.901 to 0.479; retraining on the same mask distribution and pooling three such seeds returns a nominal 0.905 C/O and 0.891 overall under the full deployment path, still in simulation (n=1024, same 105 masks), at 0.21 to 0.31 of the prior width. On 105 JWST/G395H spectral products of 13 late-T/Y dwarfs, single marginalized networks bracket the Hood et al. (2024) T8 benchmark C/O where the control excludes it, a consistency check rather than a validation, at 1.6-2.6x wider intervals.
