Matches in SemOpenAlex for { <https://semopenalex.org/work/W3177820323> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W3177820323 abstract "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." @default.
- W3177820323 created "2021-07-19" @default.
- W3177820323 creator A5050309898 @default.
- W3177820323 creator A5050531002 @default.
- W3177820323 creator A5071696590 @default.
- W3177820323 creator A5080268373 @default.
- W3177820323 date "2021-07-15" @default.
- W3177820323 modified "2023-09-27" @default.
- W3177820323 title "Lossless, Scalable Implicit Likelihood Inference for Cosmological Fields" @default.
- W3177820323 hasPublicationYear "2021" @default.
- W3177820323 type Work @default.
- W3177820323 sameAs 3177820323 @default.
- W3177820323 citedByCount "0" @default.
- W3177820323 crossrefType "posted-content" @default.
- W3177820323 hasAuthorship W3177820323A5050309898 @default.
- W3177820323 hasAuthorship W3177820323A5050531002 @default.
- W3177820323 hasAuthorship W3177820323A5071696590 @default.
- W3177820323 hasAuthorship W3177820323A5080268373 @default.
- W3177820323 hasConcept C105795698 @default.
- W3177820323 hasConcept C107673813 @default.
- W3177820323 hasConcept C11413529 @default.
- W3177820323 hasConcept C121332964 @default.
- W3177820323 hasConcept C154945302 @default.
- W3177820323 hasConcept C160234255 @default.
- W3177820323 hasConcept C162376815 @default.
- W3177820323 hasConcept C163716315 @default.
- W3177820323 hasConcept C178650346 @default.
- W3177820323 hasConcept C2776214188 @default.
- W3177820323 hasConcept C33923547 @default.
- W3177820323 hasConcept C41008148 @default.
- W3177820323 hasConcept C62520636 @default.
- W3177820323 hasConcept C95167961 @default.
- W3177820323 hasConceptScore W3177820323C105795698 @default.
- W3177820323 hasConceptScore W3177820323C107673813 @default.
- W3177820323 hasConceptScore W3177820323C11413529 @default.
- W3177820323 hasConceptScore W3177820323C121332964 @default.
- W3177820323 hasConceptScore W3177820323C154945302 @default.
- W3177820323 hasConceptScore W3177820323C160234255 @default.
- W3177820323 hasConceptScore W3177820323C162376815 @default.
- W3177820323 hasConceptScore W3177820323C163716315 @default.
- W3177820323 hasConceptScore W3177820323C178650346 @default.
- W3177820323 hasConceptScore W3177820323C2776214188 @default.
- W3177820323 hasConceptScore W3177820323C33923547 @default.
- W3177820323 hasConceptScore W3177820323C41008148 @default.
- W3177820323 hasConceptScore W3177820323C62520636 @default.
- W3177820323 hasConceptScore W3177820323C95167961 @default.
- W3177820323 hasLocation W31778203231 @default.
- W3177820323 hasOpenAccess W3177820323 @default.
- W3177820323 hasPrimaryLocation W31778203231 @default.
- W3177820323 hasRelatedWork W2532488433 @default.
- W3177820323 hasRelatedWork W2551133224 @default.
- W3177820323 hasRelatedWork W2555699918 @default.
- W3177820323 hasRelatedWork W2559982436 @default.
- W3177820323 hasRelatedWork W2728803307 @default.
- W3177820323 hasRelatedWork W2785838999 @default.
- W3177820323 hasRelatedWork W2791995286 @default.
- W3177820323 hasRelatedWork W2803889050 @default.
- W3177820323 hasRelatedWork W2889608868 @default.
- W3177820323 hasRelatedWork W2891148009 @default.
- W3177820323 hasRelatedWork W2923635089 @default.
- W3177820323 hasRelatedWork W2950774935 @default.
- W3177820323 hasRelatedWork W2951838364 @default.
- W3177820323 hasRelatedWork W2959696098 @default.
- W3177820323 hasRelatedWork W2963841639 @default.
- W3177820323 hasRelatedWork W3022346722 @default.
- W3177820323 hasRelatedWork W3033436096 @default.
- W3177820323 hasRelatedWork W3104787172 @default.
- W3177820323 hasRelatedWork W3111366808 @default.
- W3177820323 hasRelatedWork W3211865872 @default.
- W3177820323 isParatext "false" @default.
- W3177820323 isRetracted "false" @default.
- W3177820323 magId "3177820323" @default.
- W3177820323 workType "article" @default.