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.
A Framework for Obtaining Accurate Posteriors of Strong Gravitational Lensing Parameters with Flexible Priors and Implicit Likelihoods using Density Estimation
Set-based Implicit Likelihood Inference of Galaxy Cluster Mass
McWilliams Center for Cosmology and Astrophysics, Department of Physics, Carnegie Mellon University(麦卡威廉宇宙学与天体物理学中心,物理系,卡内基梅隆大学);Entertainment Technology Center, Carnegie Mellon University(娱乐技术中心,卡内基梅隆大学);Kavli IPMU (WPI), UTIAS, The University of Tokyo(Kavli IPMU(WPI),UTIAS,东京大学);Center for Data-Driven Discovery, Kavli IPMU (WPI), Japan(数据驱动发现中心,Kavli IPMU(WPI),日本)
提出结合 Deep Sets 与条件归一化流的集合式框架,利用投影星系动力学推断星系团质量后验,降低离散度并给出可靠不确定性。
LIMFAST. IV. Learning High-Redshift Galaxy Formation from Multiline Intensity Mapping with Implicit Likelihood Inference
Northwestern University(西北大学);NSF--Simons AI Institute for the Sky (SkAI)(美国国家科学基金会-西蒙斯天空人工智能研究所);University of Pennsylvania(宾夕法尼亚大学);Jet Propulsion Laboratory, California Institute of Technology(加州理工学院喷气推进实验室);California Institute of Technology(加州理工学院);University of California, Los Angeles(加利福尼亚大学洛杉矶分校)
Forecasting Dark Matter Subhalo Constraints from Stellar Streams using Implicit Likelihood Inference
Northwestern University(西北大学);NSF-Simons AI Institute for the Sky(NSF-西蒙斯天空AI研究所);Carnegie Mellon University(卡内基梅隆大学);University of Washington(华盛顿大学);University of Chicago(芝加哥大学);Fermi National Accelerator Laboratory(费米国家加速器实验室);University of Surrey(萨里大学);Newcastle University(纽卡斯尔大学);Lund University(隆德大学)
Bayesian cosmological inference through implicit cross-correlation statistics
对下一代星系数据的分析需要准确处理系统效应,如观测星系与底层物质密度场之间的偏差。然而,所提出的模型的现象是数值昂贵或太不准确,以实现无偏的宇宙学参数的推断,即使在轻度非线性尺度的数据。作为一种替代构建准确的星系偏差模型,需要了解星系形成,我们建议构建似然分布的贝叶斯正向建模方法是不敏感的线性,尺度依赖的偏见,并提供对模型误指定的鲁棒性。我们使用最大熵参数来构造仅考虑数据和推断量之间的相关性的似然分布。通过设计,这些相关性对线性星系偏置关系不敏感,提供了所需的鲁棒性。该方法的实施和测试中的马尔可夫链蒙特卡罗方法。该方法是评估使用晕模拟目录的基础上标准的完整,宇宙学,N体模拟。我们得到公正的和严格的约束宇宙学参数只利用线性互相关率为$k\le 0.10$兆秒钟/小时。对质量约为10 $^{12}$ M $_\odot$到约10 $^{13}$ M $_\odot$的晕的测试表明,在获得宇宙学的鲁棒约束时,可以忽略线性的、尺度相关的、偏置函数的所有细节。我们的研究结果提供了一个很有前途的路径向前分析星系调查,而不需要准确地模拟星系偏置的细节,但通过设计强大的似然推理。