LINNA: Likelihood Inference Neural Network Accelerator

Abstract

Bayesian posterior inference of modern multi-probe cosmological analyses incurs massive computational costs. For instance, depending on the combinations of probes, a single posterior inference for the Dark Energy Survey (DES) data had a wall-clock time that ranged from 1 to 21 days using a state-of-the-art computing cluster with 100 cores. These computational costs have severe environmental impacts and the long wall-clock time slows scientific productivity. To address these difficulties, we introduce LINNA: the Likelihood Inference Neural Network Accelerator. Relative to the baseline DES analyses, LINNA reduces the computational cost associated with posterior inference by a factor of 8--50. If applied to the first-year cosmological analysis of Rubin Observatory's Legacy Survey of Space and Time (LSST Y1), we conservatively estimate that LINNA will save more than US $\$300,000$ on energy costs, while simultaneously reducing ${\mathrm{CO}}_2$ emission by $2,400$ tons. To accomplish these reductions, LINNA automatically builds training data sets, creates neural network surrogate models, and produces a Markov chain that samples the posterior. We explicitly verify that LINNA accurately reproduces the first-year DES (DES Y1) cosmological constraints derived from a variety of different data vectors with our default code settings, without needing to retune the algorithm every time. Further, we find that LINNA is sufficient for enabling accurate and efficient sampling for LSST Y10 multi-probe analyses. We make LINNA publicly available at https://github.com/chto/linna, to enable others to perform fast and accurate posterior inference in contemporary cosmological analyses.

21 pages, 12 figures, submitted to JCAP, comments are welcome

References in corpus (11)

• The NumPy array: a structure for efficient numerical computation
• Dark Energy Survey Year 3 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• KiDS-450 + 2dFLenS: Cosmological parameter constraints from weak gravitational lensing tomography and overlapping redshift-space galaxy clustering
• Fast likelihood-free cosmology with neural density estimators and active learning
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• Fast cosmological parameter estimation using neural networks
• zeus: A Python implementation of Ensemble Slice Sampling for efficient Bayesian parameter inference
• {\sc CosmoNet}: fast cosmological parameter estimation in non-flat models using neural networks
• PkANN - I. Non-linear matter power spectrum interpolation through artificial neural networks
• Kernel-Based Emulator for the 3D Matter Power Spectrum from CLASS
• Neural Network Acceleration of Large-scale Structure Theory Calculations

Cited by in corpus (10)

• Cosmological Constraints from Galaxy Clusters and Groups in the eROSITA Final Equatorial Depth Survey
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Capse.jl: efficient and auto-differentiable CMB power spectra emulation
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• Fast and effortless computation of profile likelihoods using CONNECT
• CombineHarvesterFlow: Joint Probe Analysis Made Easy with Normalizing Flows
• Faster Bayesian inference with neural network bundles and new results for $f(R)$ models
• Informed total-error-minimizing priors: Interpretable cosmological parameter constraints despite complex nuisance effects
• Environmental sustainability in basic research: a perspective from HECAP+
• Variational autoencoder for generating realistic $N$-body simulations for dark matter halos