Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference

Abstract

We combine amortized neural posterior estimation with importance sampling for fast and accurate gravitational-wave inference. We first generate a rapid proposal for the Bayesian posterior using neural networks, and then attach importance weights based on the underlying likelihood and prior. This provides (1) a corrected posterior free from network inaccuracies, (2) a performance diagnostic (the sample efficiency) for assessing the proposal and identifying failure cases, and (3) an unbiased estimate of the Bayesian evidence. By establishing this independent verification and correction mechanism we address some of the most frequent criticisms against deep learning for scientific inference. We carry out a large study analyzing 42 binary black hole mergers observed by LIGO and Virgo with the SEOBNRv4PHM and IMRPhenomXPHM waveform models. This shows a median sample efficiency of $\approx 10\%$ (two orders-of-magnitude better than standard samplers) as well as a ten-fold reduction in the statistical uncertainty in the log evidence. Given these advantages, we expect a significant impact on gravitational-wave inference, and for this approach to serve as a paradigm for harnessing deep learning methods in scientific applications.

8+7 pages, 1+5 figures. [v2]: Minor updates to match published version, code available at https://github.com/dingo-gw/dingo

References in corpus (7)

• Advanced Virgo: a 2nd generation interferometric gravitational wave detector
• Advanced LIGO
• Robust parameter estimation for compact binaries with ground-based gravitational-wave observations using the LALInference software library
• Multipolar Effective-One-Body Waveforms for Precessing Binary Black Holes: Construction and Validation
• The curious case of GW200129: interplay between spin-precession inference and data-quality issues
• Adapting to noise distribution shifts in flow-based gravitational-wave inference
• The False Alarms induced by Gaussian Noise in Gravitational Wave Detectors

Cited by in corpus (64)

• SEOBNRv5PHM: Next generation of accurate and efficient multipolar precessing-spin effective-one-body waveforms for binary black holes
• Laying the foundation of the effective-one-body waveform models SEOBNRv5: improved accuracy and efficiency for spinning non-precessing binary black holes
• The Science of the Einstein Telescope
• Real-time gravitational-wave inference for binary neutron stars using machine learning
• Neural Posterior Estimation with guaranteed exact coverage: the ringdown of GW150914
• Evidence for eccentricity in the population of binary black holes observed by LIGO-Virgo-KAGRA
• VARAHA: A Fast Non-Markovian sampler for estimating Gravitational-Wave posteriors
• Gravitational wave populations and cosmology with neural posterior estimation
• New Gravitational Wave Discoveries Enabled by Machine Learning
• Fast likelihood evaluation using meshfree approximations for reconstructing compact binary sources
• Adapting to noise distribution shifts in flow-based gravitational-wave inference
• GW200208_222617 as an eccentric black-hole binary merger: properties and astrophysical implications
• Applications of machine learning in gravitational wave research with current interferometric detectors
• Advancing Space-Based Gravitational Wave Astronomy: Rapid Parameter Estimation via Normalizing Flows
• Eccentricity signatures in LIGO-Virgo-KAGRA's BNS and NSBH binaries
• Analysis of GWTC-3 with fully precessing numerical relativity surrogate models
• Bayesian evidence estimation from posterior samples with normalizing flows
• Neural post-Einsteinian framework for efficient theory-agnostic tests of general relativity with gravitational waves
• Simulation-based inference of black hole ringdowns in the time domain
• Comparative study of 1D and 2D convolutional neural network models with attribution analysis for gravitational wave detection from compact binary coalescences
• Rapid inference and comparison of gravitational-wave population models with neural variational posteriors
• Decoding Long-duration Gravitational Waves from Binary Neutron Stars with Machine Learning: Parameter Estimation and Equations of State
• Robust inference of gravitational wave source parameters in the presence of noise transients using normalizing flows
• Parameter estimation of microlensed gravitational waves with Conditional Variational Autoencoders
• Rapid identification of time-frequency domain gravitational wave signals from binary black holes using deep learning
• Classifying binary black holes from Population III stars with the Einstein Telescope: A machine-learning approach
• Mass and tidal parameter extraction from gravitational waves of binary neutron stars mergers using deep learning
• Flow Matching for Atmospheric Retrieval of Exoplanets: Where Reliability meets Adaptive Noise Levels
• Two-Step Procedure to Detect Cosmological Gravitational Wave Backgrounds with Next-Generation Terrestrial Gravitational-Wave Detectors
• Deep learning to detect gravitational waves from binary close encounters: Fast parameter estimation using normalizing flows
• Waveform systematics in identifying strongly gravitationally lensed gravitational waves: Posterior overlap method
• Searching for gravitational waves from stellar-mass binary black holes early inspiral
• Time delay interferometry with minimal null frequencies
• Reconstruction of binary black hole harmonics in LIGO using deep learning
• Deep-Learning Classification and Parameter Inference of Rotational Core-Collapse Supernovae
• Inferring Binary Properties from Gravitational Wave Signals
• Navigating Unknowns: Deep Learning Robustness for Gravitational Wave Signal Reconstruction
• Modeling matter(s) in SEOBNRv5THM: Generating fast and accurate effective-one-body waveforms for spin-aligned binary neutron stars
• Accelerated parameter estimation of supermassive black hole binaries in LISA using a meshfree approximation
• Accelerating Bayesian Sampling for Massive Black Hole Binaries with Prior Constraints from Conditional Variational Autoencoder
• Next-generation global gravitational-wave detector network: Impact of detector orientation on compact binary coalescence and stochastic gravitational-wave background searches
• Fast and accurate parameter estimation of high-redshift sources with the Einstein Telescope
• Improving the detection sensitivity to primordial stochastic gravitational waves with reduced astrophysical foregrounds. II. Subthreshold binary neutron stars
• Compact Binary Coalescence Gravitational Wave Signals Counting and Separation
• Comparison of neural network architectures for feature extraction from binary black hole merger waveforms
• Accelerating LISA inference with Gaussian processes
• Robust, Rapid, and Simple Gravitational-wave Parameter Estimation
• Accelerated inference of microlensed gravitational waves with machine learning
• Extracting overlapping gravitational-wave signals of galactic compact binaries: a mini review
• Rapid Parameter Estimation for Merging Massive Black Hole Binaries Using Continuous Normalizing Flows
• Discovering gravitational waveform distortions from lensing: A deep dive into GW231123
• A practical Bayesian method for gravitational-wave ringdown analysis with multiple modes
• Comparing next-generation detector configurations for high-redshift gravitational wave sources with neural posterior estimation
• Flexible Gravitational-Wave Parameter Estimation with Transformers
• Parameter inference of millilensed gravitational waves using neural spline flows
• labrador: A domain-optimized machine-learning tool for gravitational wave inference
• A Robust and Efficient F-statistic-based Framework for Consistent Bayesian Inference of Compact Binary Coalescences
• Extract non-Gaussian Features in Gravitational Wave Observation Data Using Self-Supervised Learning
• Conditional variational autoencoders for cosmological model discrimination and anomaly detection in cosmic microwave background power spectra
• Auto-encoder model for faster generation of effective one-body gravitational waveform approximations
• Model-agnostic search of gravitational wave echoes in LVK data
• Combining simulation-based inference and universal relations for precise and accurate neutron star science
• Accurate and efficient simulation-based inference for massive black-hole binaries with LISA
• Assessment of normalizing flows for parameter estimation on time-frequency representations of gravitational-wave data