arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.07405astro-ph.CO
收藏

宇宙学场的无损可扩展隐式似然推断

Lossless, Scalable Implicit Likelihood Inference for Cosmological Fields

T. Lucas Makinen, Tom Charnock, Justin Alsing, Benjamin D. Wandelt

摘要:

我们提出了一个比较基于模拟的推理,以充分,基于场的宇宙数据分析的分析推理。要做到这一点,我们探讨两种情况下的信息内容是可计算的参数推断分析:高斯随机场的协方差取决于参数通过功率谱;和相关的对数正态场与宇宙功率谱。我们比较两种推理技术:i)使用已知似然的显式字段级推理和ii)通过信息最大化神经网络(IMNN)压缩的最大信息汇总统计量的隐式似然推理。我们发现,a)从卷积神经网络压缩获得的摘要不会丢失信息,因此对于高斯协方差和对数正态情况都使已知的字段信息内容饱和,b)使用这些信息量最大的非线性摘要的基于模拟的推断几乎可以恢复字段级推断的精确后验,绕过评估昂贵的似然或逆协方差矩阵的需要,以及c)即使对于这个简单的例子,隐式的、基于模拟的似然性也比利用显式似然性的推断引起小得多的计算成本。这项工作使用了一个新的IMNNs实现在$\texttt{Jax}$,可以充分利用可微分的模拟和推理流水线。我们还证明了一个单一的再训练的IMNN摘要有效地实现了理论上的最大信息,增强了鲁棒性的基准模型的选择,其中IMNN的训练。

英文摘要:

We present a comparison of simulation-based inference to full, field-based analytical inference in cosmological data analysis. To do so, we explore parameter inference for two cases where the information content is calculable analytically: Gaussian random fields whose covariance depends on parameters through the power spectrum; and correlated lognormal fields with cosmological power spectra. We compare two inference techniques: i) explicit field-level inference using the known likelihood and ii) implicit likelihood inference with maximally informative summary statistics compressed via Information Maximising Neural Networks (IMNNs). We find that a) summaries obtained from convolutional neural network compression do not lose information and therefore saturate the known field information content, both for the Gaussian covariance and the lognormal cases, b) simulation-based inference using these maximally informative nonlinear summaries recovers nearly losslessly the exact posteriors of field-level inference, bypassing the need to evaluate expensive likelihoods or invert covariance matrices, and c) even for this simple example, implicit, simulation-based likelihood incurs a much smaller computational cost than inference with an explicit likelihood. This work uses a new IMNNs implementation in $\texttt{Jax}$ that can take advantage of fully-differentiable simulation and inference pipeline. We also demonstrate that a single retraining of the IMNN summaries effectively achieves the theoretically maximal information, enhancing the robustness to the choice of fiducial model where the IMNN is trained.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑