Bayesian evidence estimation from posterior samples with normalizing flows

Abstract

We propose a novel method ($floZ$), based on normalizing flows, to estimate the Bayesian evidence (and its numerical uncertainty) from a pre-existing set of samples drawn from the unnormalized posterior distribution. We validate it on distributions whose evidence is known analytically, up to 15 parameter space dimensions, and compare with two state-of-the-art techniques for estimating the evidence: nested sampling (which computes the evidence as its main target) and a $k$-nearest-neighbors technique that produces evidence estimates from posterior samples. Provided representative samples from the target posterior are available, our method is more robust to posterior distributions with sharp features, especially in higher dimensions. For a simple multivariate Gaussian, we demonstrate its accuracy for up to 200 dimensions with $10^5$ posterior samples. $floZ$ has wide applicability, e.g., to estimate evidence from variational inference, Markov Chain Monte Carlo samples, or any other method that delivers samples and their likelihood from the unnormalized posterior density. As a physical application, we use $floZ$ to compute the Bayes factor for the presence of the first overtone in the ringdown signal of the gravitational wave data of GW150914, finding good agreement with nested sampling.

15 pages, 8 figures, 1 table

References in corpus (19)

• The NANOGrav 15-year Data Set: Evidence for a Gravitational-Wave Background
• Bayes in the sky: Bayesian inference and model selection in cosmology
• Search for an isotropic gravitational-wave background with the Parkes Pulsar Timing Array
• Searching for the nano-Hertz stochastic gravitational wave background with the Chinese Pulsar Timing Array Data Release I
• The second data release from the European Pulsar Timing Array III. Search for gravitational wave signals
• Observational Black Hole Spectroscopy: A time-domain multimode analysis of GW150914
• Analysis of Ringdown Overtones in GW150914
• Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference
• The Indian Pulsar Timing Array: First data release
• Using rational filters to uncover the first ringdown overtone in GW150914
• Neural Posterior Estimation with guaranteed exact coverage: the ringdown of GW150914
• Comment on "Analysis of Ringdown Overtones in GW150914''
• Reply to Comment on "Analysis of Ringdown Overtones in GW150914"
• Retrieval Study of Brown Dwarfs Across the L-T Sequence
• Gravitational wave populations and cosmology with neural posterior estimation
• Adapting to noise distribution shifts in flow-based gravitational-wave inference
• Bayesian model comparison for simulation-based inference
• Effect of Noise Estimation in Time-Domain Ringdown Analysis: A Case Study with GW150914
• Evidence Networks: simple losses for fast, amortized, neural Bayesian model comparison

Cited by in corpus (7)

• Impact of relativistic waveforms in LISA's science objectives with extreme-mass-ratio inspirals
• Stellar flare morphology with TESS across the main sequence
• Reanalyzing the ringdown signal of GW150914 using the F-statistic method
• Enhancing Cosmological Model Selection with Interpretable Machine Learning
• A flexible parameterization to test early physics solutions to the Hubble tension with future CMB data
• The Velocity Field Olympics: Assessing velocity field reconstructions with direct distance tracers
• Calibrating Bayesian Tension Statistics using Neural Ratio Estimation