AI 中文总结
本文利用w_iCDM模型，通过LtU-ILI流程结合多组宇宙学数据，经6轮共12万次模拟训练神经网络重建暗能量w(z)，发现多数分箱结果与宇宙学常数一致，首个分箱略倾向动力学暗能量，末两个分箱未受约束。
AI 中文摘要
本文为利用红移分箱法重建暗能量（DE）的状态方程（EOS）w(z)，首先引入了在7个红移分箱中采用分段常数EOS的w_iCDM模型。随后，采用Learning the Universe隐似然推断（LtU-ILI）流程，对w_i开展多轮隐似然推断，所用宇宙学数据组合包括普朗克2018的TT、TE、EE及透镜功率谱，DESI DR2的距离比率，以及Pantheon+样本的Ia型超新星（SNIa）校正视星等。更具体地，我们通过CLASS构建宇宙微波背景（CMB）功率谱、重子声学振荡（BAO）距离比率和Ia型超新星视星等模拟器，并将其嵌入LtU-ILI流程。采用序列神经似然估计（SNLE），我们分6轮共6×20000次模拟来训练神经网络，以拟合w_iCDM正向模型的“黑箱”似然。最后，结合w_i的估计后验分布，我们发现除最后两个分箱的w_5和w_6未受约束外，w(z)的重建结果在第一个分箱中略微倾向于动力学暗能量，在其余分箱中与宇宙学常数在68%置信水平（C.L.）下一致。
英文摘要
In this paper, to reconstruct the equation of state (EOS) of dark energy (DE) $w(z)$ with the redshift binning method, we first introduce a $w_i$CDM model with a piecewise-constant EOS in $7$ redshift bins. Then, we turn to the Learning the Universe Implicit Likelihood Inference (LtU-ILI) pipeline to perform a multi-round ILI of $w_i$ from the cosmological data combination, including $TT$, $TE$, $EE$ and lensing power spectra of Planck 2018, distance ratios of DESI DR2 and corrected apparent magnitudes of SNIa from Pantheon+ sample. More precisely, we build the Cosmic Microwave Background (CMB) power spectrum, Baryon Acoustic Oscillation (BAO) distance ratio and Type Ia Supernovae (SNIa) apparent magnitude simulators by $\mathtt{CLASS}$ and embed them into the LtU-ILI pipeline. And, using Sequential Neural Likelihood Estimation (SNLE), we sequentially train neural networks with $6$ rounds of total $6\times20000$ simulations to target a ``black box'' likelihood of our forward model $w_i$CDM. Finally, with the estimated posteriors of $w_i$, we find that except for the unconstrained $w_5$ and $w_6$ (the last two bins), our reconstruction of $w(z)$ marginally favors dynamical DE in the first bin and is consistent with the cosmological constant at $68\%$ C.L. in the other bins.
Comments11 pages, 6 figures
