Deep Learning and genetic algorithms for cosmological Bayesian inference speed-up

Abstract

In this paper, we present a novel approach to accelerate the Bayesian inference process, focusing specifically on the nested sampling algorithms. Bayesian inference plays a crucial role in cosmological parameter estimation, providing a robust framework for extracting theoretical insights from observational data. However, its computational demands can be substantial, primarily due to the need for numerous likelihood function evaluations. Our proposed method utilizes the power of deep learning, employing feedforward neural networks to approximate the likelihood function dynamically during the Bayesian inference process. Unlike traditional approaches, our method trains neural networks on-the-fly using the current set of live points as training data, without the need for pre-training. This flexibility enables adaptation to various theoretical models and datasets. We perform simple hyperparameter optimization using genetic algorithms to suggest initial neural network architectures for learning each likelihood function. Once sufficient accuracy is achieved, the neural network replaces the original likelihood function. The implementation integrates with nested sampling algorithms and has been thoroughly evaluated using both simple cosmological dark energy models and diverse observational datasets. Additionally, we explore the potential of genetic algorithms for generating initial live points within nested sampling inference, opening up new avenues for enhancing the efficiency and effectiveness of Bayesian inference methods.

16 pages, 4 figures; matches the version published in Physical Review D

References in corpus (51)

• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample
• MultiNest: an efficient and robust Bayesian inference tool for cosmology and particle physics
• The Complete Light-curve Sample of Spectroscopically Confirmed Type Ia Supernovae from Pan-STARRS1 and Cosmological Constraints from The Combined Pantheon Sample
• The 6dF Galaxy Survey: Baryon Acoustic Oscillations and the Local Hubble Constant
• Exploring the Expansion History of the Universe
• Planck 2018 results. X. Constraints on inflation
• dynesty: A Dynamic Nested Sampling Package for Estimating Bayesian Posteriors and Evidences
• The Clustering of the SDSS DR7 Main Galaxy Sample I: A 4 per cent Distance Measure at z=0.15
• X-ray spectral modelling of the AGN obscuring region in the CDFS: Bayesian model selection and catalogue
• Constraints on the redshift dependence of the dark energy potential
• A 6% measurement of the Hubble parameter at $z\sim 0.45$: direct evidence of the epoch of cosmic re-acceleration
• Cosmic Chronometers: Constraining the Equation of State of Dark Energy. I: H(z) Measurements
• Information criteria for astrophysical model selection
• Raising the bar: new constraints on the Hubble parameter with cosmic chronometers at z$\sim$2
• Conservative Constraints on Early Cosmology: an illustration of the Monte Python cosmological parameter inference code
• Cosmological implications of baryon acoustic oscillation (BAO) measurements
• Age-dating Luminous Red Galaxies observed with the Southern African Large Telescope
• Efficient sampling of fast and slow cosmological parameters
• PolyChord: nested sampling for cosmology
• Constraints on the equation of state of dark energy and the Hubble constant from stellar ages and the CMB
• The clustering of the SDSS-IV extended Baryon Oscillation Spectroscopic Survey DR14 quasar sample: First measurement of Baryon Acoustic Oscillations between redshift 0.8 and 2.2
• A Nested Sampling Algorithm for Cosmological Model Selection
• Data analysis recipes: Using Markov Chain Monte Carlo
• New constraints on cosmological parameters and neutrino properties using the expansion rate of the Universe to z~1.75
• Fast and reliable MCMC for cosmological parameter estimation
• Baryon acoustic oscillations from the cross-correlation of Ly$\alpha$ absorption and quasars in eBOSS DR14
• Baryon acoustic oscillations at z = 2.34 from the correlations of Ly$\alpha$ absorption in eBOSS DR14
• Fast likelihood-free cosmology with neural density estimators and active learning
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• DIAMONDS: a new Bayesian Nested Sampling tool. Application to Peak Bagging of solar-like oscillations
• Testing General Relativity using Bayesian model selection: Applications to observations of gravitational waves from compact binary systems
• SKYNET: an efficient and robust neural network training tool for machine learning in astronomy
• Use of the MultiNest algorithm for gravitational wave data analysis
• Approximate Bayesian Computation for Forward Modeling in Cosmology
• Internal Robustness of Growth Rate data
• cosmoabc: Likelihood-free inference via Population Monte Carlo Approximate Bayesian Computation
• Fast cosmological parameter estimation using neural networks
• astroABC: An Approximate Bayesian Computation Sequential Monte Carlo sampler for cosmological parameter estimation
• Parameters Estimation for the Cosmic Microwave Background with Bayesian Neural Networks
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Copula Cosmology: Constructing a Likelihood Function
• BAMBI: blind accelerated multimodal Bayesian inference
• Cosmological parameter estimation via iterative emulation of likelihoods
• Neural network reconstructions for the Hubble parameter, growth rate and distance modulus
• Accelerated Bayesian inference using deep learning
• Observational cosmology with Artificial Neural Networks
• Cosmological parameter estimation with Genetic Algorithms
• Gaussianisation for fast and accurate inference from cosmological data
• Neural Networks Optimized by Genetic Algorithms in Cosmology
• A new code for parameter estimation in searches for gravitational waves from known pulsars
• Faster Bayesian inference with neural network bundles and new results for $f(R)$ models

Cited by in corpus (6)

• The CosmoVerse White Paper: Addressing observational tensions in cosmology with systematics and fundamental physics
• Non-parametric reconstruction of cosmological observables using Gaussian Processes Regression
• Anisotropic cosmology using observational datasets: exploring via machine learning approaches
• A Quantum Genetic Algorithm with application to Cosmological Parameters Estimation
• Variational autoencoder for generating realistic $N$-body simulations for dark matter halos
• Nature-inspired optimization, the Philippine Eagle, and cosmological parameter estimation