Accelerated Bayesian inference using deep learning

Abstract

We present a novel Bayesian inference tool that uses a neural network to parameterise efficient Markov Chain Monte-Carlo (MCMC) proposals. The target distribution is first transformed into a diagonal, unit variance Gaussian by a series of non-linear, invertible, and non-volume preserving flows. Neural networks are extremely expressive, and can transform complex targets to a simple latent representation. Efficient proposals can then be made in this space, and we demonstrate a high degree of mixing on several challenging distributions. Parameter space can naturally be split into a block diagonal speed hierarchy, allowing for fast exploration of subspaces where it is inexpensive to evaluate the likelihood. Using this method, we develop a nested MCMC sampler to perform Bayesian inference and model comparison, finding excellent performance on highly curved and multi-modal analytic likelihoods. We also test it on {\em Planck} 2015 data, showing accurate parameter constraints, and calculate the evidence for simple one-parameter extensions to LCDM in $\sim 20$ dimensional parameter space. Our method has wide applicability to a range of problems in astronomy and cosmology.

12 pages, 6 figures

References in corpus (2)

• No evidence for extensions to the standard cosmological model
• A-NICE-MC: Adversarial Training for MCMC

Cited by in corpus (19)

• Nested sampling for physical scientists
• Nested Sampling with Normalising Flows for Gravitational-Wave Inference
• Nested Sampling Methods
• A deep learning approach to cosmological dark energy models
• Bilby-MCMC: An MCMC sampler for gravitational-wave inference
• Accelerating astronomical and cosmological inference with Preconditioned Monte Carlo
• Non-Gaussian estimates of tensions in cosmological parameters
• Importance nested sampling with normalising flows
• A High-Resolution Investigation of the Multi-Phase ISM in a Galaxy during the First Two Billion Years
• Observational cosmology with Artificial Neural Networks
• Nested sampling with any prior you like
• Bayesian evidence estimation from posterior samples with normalizing flows
• Detection of Dipole Modulation in CMB Temperature Anisotropy Maps from WMAP and Planck using Artificial Intelligence
• Deep Learning and genetic algorithms for cosmological Bayesian inference speed-up
• Estimation of Full Sky Power Spectrum between Intermediate to Large Angular Scales from Partial Sky CMB Anisotropies using Artificial Neural Network
• What does a cosmological experiment really measure? Covariant posterior decomposition with normalizing flows
• Faster Bayesian inference with neural network bundles and new results for $f(R)$ models
• ParamANN: A Neural Network to Estimate Cosmological Parameters for $\Lambda$CDM Universe Using Hubble Measurements
• Reconstruction of full sky CMB $\mathbf{E}$ and $\mathbf{B}$ modes spectra removing $\mathbf{E}$-to-$\mathbf{B}$ leakage from partial sky using deep learning