标题： DeepCHART：使用深度学习从Ly$α$森林调查中绘制暗物质密度场的三维分布 显示英文标题

标题： DeepCHART: Mapping the 3D dark matter density field from Ly$α$ forest surveys using deep learning

摘要： 我们提出 DeepCHART（通过断层扫描进行宇宙异质性和天体物理重建的深度学习），这是一个深度学习框架，旨在从 Ly$\alpha$森林光谱中重建红移$z=2.5$处的三维暗物质密度场。 利用具有 U-Net 架构的 3D 变分自编码器，DeepCHART 进行快速、无需似然函数的推断，准确捕捉宇宙流体动力学模拟中嵌入的非线性引力动力学和重子过程。 当应用于结合 Ly$\alpha$森林吸收和同时期星系位置的联合数据集时，重建质量进一步提高。 对于当前的调查，如 Subaru/PFS、CLAMATO 和 LATIS，平均横向视线间距为$d_\perp=2.4h^{-1}$cMpc，DeepCHART 在密度范围$0.4<\Delta_{\rm DM}<15$内实现了高保真重建，体素级皮尔逊相关系数为$\rho\simeq 0.77$。 这些重建数据是使用 Ly$\alpha$ 森林谱获得的，信噪比低至 2，仪器分辨率 $R=2500$，符合 Subaru/PFS 的规格。 对于未来由 ELT/MOSAIC 和 WST/IFS 等仪器（分辨率 $d_\perp\simeq 1h^{-1}\mathrm{cMpc}$）支持的高密度巡天，相关性将在更宽的动态范围（$0.25<\Delta_{\rm DM} <40$）内提升至 $\rho\simeq 0.90$。 该框架能够可靠地恢复暗物质密度概率密度函数 (PDF) 以及功率谱，仅在中等尺度上存在轻微的抑制。 在宇宙网分类方面， DeepCHART 成功识别了 81% 的空洞、75% 的片状结构、63% 的细丝和 43% 的节点。 我们提出 DeepCHART 作为一个强大且可扩展的场级宇宙学推断框架，可轻松推广到其他可观测量，并提供一种强大而有效的方法来最大化即将进行的光谱调查的科学回报。 显示英文摘要

摘要： We present DeepCHART (Deep learning for Cosmological Heterogeneity and Astrophysical Reconstruction via Tomography), a deep learning framework designed to reconstruct the three-dimensional dark matter density field at redshift $z=2.5$ from Ly$\alpha$ forest spectra. Leveraging a 3D variational autoencoder with a U-Net architecture, DeepCHART performs fast, likelihood-free inference, accurately capturing the non-linear gravitational dynamics and baryonic processes embedded in cosmological hydrodynamical simulations. When applied to joint datasets combining Ly$\alpha$ forest absorption and coeval galaxy positions, the reconstruction quality improves further. For current surveys, such as Subaru/PFS, CLAMATO, and LATIS, with an average transverse sightline spacing of $d_\perp=2.4h^{-1}$cMpc, DeepCHART achieves high-fidelity reconstructions over the density range $0.4<\Delta_{\rm DM}<15$, with a voxel-wise Pearson correlation coefficient of $\rho\simeq 0.77$. These reconstructions are obtained using Ly$\alpha$ forest spectra with signal-to-noise ratios as low as 2 and instrumental resolution $R=2500$, matching Subaru/PFS specifications. For future high-density surveys enabled by instruments such as ELT/MOSAIC and WST/IFS with $d_\perp\simeq 1h^{-1}\mathrm{cMpc}$, the correlation improves to $\rho\simeq 0.90$ across a wider dynamic range ($0.25<\Delta_{\rm DM}<40$). The framework reliably recovers the dark matter density PDF as well as the power spectrum, with only mild suppression at intermediate scales. In terms of cosmic web classification, DeepCHART successfully identifies 81% of voids, 75% of sheets, 63% of filaments, and 43% of nodes. We propose DeepCHART as a powerful and scalable framework for field-level cosmological inference, readily generalisable to other observables, and offering a robust, efficient means of maximising the scientific return of upcoming spectroscopic surveys.