Hybrid summary statistics: neural weak lensing inference beyond the power spectrum
arXiv:2407.18909 · doi:10.1088/1475-7516/2025/01/095
Abstract
In inference problems, we often have domain knowledge which allows us to define summary statistics that capture most of the information content in a dataset. In this paper, we present a hybrid approach, where such physics-based summaries are augmented by a set of compressed neural summary statistics that are optimised to extract the extra information that is not captured by the predefined summaries. The resulting statistics are very powerful inputs to simulation-based or implicit inference of model parameters. We apply this generalisation of Information Maximising Neural Networks (IMNNs) to parameter constraints from tomographic weak gravitational lensing convergence maps to find summary statistics that are explicitly optimised to complement angular power spectrum estimates. We study several dark matter simulation resolutions in low- and high-noise regimes. We show that i) the information-update formalism extracts at least and up to as much information as the angular power spectrum in all noise regimes, ii) the network summaries are highly complementary to existing 2-point summaries, and iii) our formalism allows for networks with smaller, physically-informed architectures to match much larger regression networks with far fewer simulations needed to obtain asymptotically optimal inference.
16 pages, 11 figures. Submitted to JCAP. We provide publicly available code at https://github.com/tlmakinen/hybridStatsWL
References in corpus (50)
- Empirical Evaluation of Rectified Activations in Convolutional Network
- Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
- Power Spectra for Cold Dark Matter and its Variants
- KiDS-1000 Cosmology: Cosmic shear constraints and comparison between two point statistics
- Cosmology with cosmic shear observations: a review
- Dark Energy Survey Year 3 Results: Cosmology from Cosmic Shear and Robustness to Data Calibration
- Dark Energy Survey Year 3 Results: Cosmology from Cosmic Shear and Robustness to Modeling Uncertainty
- Massive Lossless Data Compression and Multiple Parameter Estimation from Galaxy Spectra
- Fast likelihood-free cosmology with neural density estimators and active learning
- Hyper Suprime-Cam Year 3 Results: Cosmology from Cosmic Shear Power Spectra
- Cosmological constraints with deep learning from KiDS-450 weak lensing maps
- A new approach to observational cosmology using the scattering transform
- Probing Cosmology with Weak Lensing Peak Counts
- Likelihood-free inference with neural compression of DES SV weak lensing map statistics
- Past and present cosmic structure in the SDSS DR7 main sample
- Automatic physical inference with information maximising neural networks
- Weak lensing cosmology with convolutional neural networks on noisy data
- Generalized massive optimal data compression
- Cosmological constraints from noisy convergence maps through deep learning
- preparation: XV. Forecasting cosmological constraints for the and CMB joint analysis
- Super-resolution emulator of cosmological simulations using deep physical models
- Cosmological parameters, shear maps and power spectra from CFHTLenS using Bayesian hierarchical inference
- Euclid Preparation. XXVIII. Forecasts for ten different higher-order weak lensing statistics
- On the insufficiency of arbitrarily precise covariance matrices: non-Gaussian weak lensing likelihoods
- Cosmological inference from Bayesian forward modelling of deep galaxy redshift surveys
- A Full CDM Analysis of KiDS-1000 Weak Lensing Maps using Deep Learning
- Re-capturing cosmic information
- Cosmological constraints with weak lensing peak counts and second-order statistics in a large-field survey
- KiDS-1000: Cosmology with improved cosmic shear measurements
- The Cosmic Graph: Optimal Information Extraction from Large-Scale Structure using Catalogues
- Bayesian forward modelling of cosmic shear data
- Cosmological information in Gaussianised weak lensing signals
- LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology
- Optimal non-linear transformations for large scale structure statistics
- KiDS-SBI: Simulation-based inference analysis of KiDS-1000 cosmic shear
- Neural physical engines for inferring the halo mass distribution function
- Sampling-Based Accuracy Testing of Posterior Estimators for General Inference
- A comparative study of cosmological constraints from weak lensing using Convolutional Neural Networks
- Cosmological constraints from weak lensing scattering transform using HSC Y1 data
- Almanac: Weak Lensing power spectra and map inference on the masked sphere
- Field-level inference of cosmic shear with intrinsic alignments and baryons
- pmwd: A Differentiable Cosmological Particle-Mesh -body Library
- Forecasting the power of Higher Order Weak Lensing Statistics with automatically differentiable simulations
- Enhancing the Cosmic Shear Power Spectrum
- : A generative, fast, and differentiable halo model for wide-field galaxy surveys
- Catalog-free modeling of galaxy types in deep images: Massive dimensional reduction with neural networks
- Almanac: MCMC-based signal extraction of power spectra and maps on the sphere
- Higher-order statistics of the large-scale structure from photometric redshifts
- Optimal Neural Summarisation for Full-Field Weak Lensing Cosmological Implicit Inference
- Unifying Summary Statistic Selection for Approximate Bayesian Computation