mlsanityData to 7 Oct 2026

Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference

arXiv 2210.05686 ↗First posted 11 Oct 2022gr-qc astro-ph.IM cs.LGcited by 28

Not on the map. Why?

Abstract (arXiv)

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 ≈ 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.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). Builds on 24 of its 74 references that are in the map. First posted on arXiv 11 Oct 2022.

Builds on

24 of 74 references.

References · 24

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2015Variational Inference with Normalizing Flowscited by 2,161
  6. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
  7. 2017Masked Autoregressive Flow for Density Estimationcited by 680
  8. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 636
  9. 2019Neural Spline Flowscited by 590
  10. 2019The frontier of simulation-based inferencecited by 490
  11. 2018Neural Importance Samplingcited by 154
  12. 2019Flow-based generative models for Markov chain Monte Carlo in lattice field theorycited by 147
12 more
  1. 2020Equivariant flow-based sampling for lattice gauge theorycited by 102
  2. 2021Real-time gravitational-wave science with neural posterior estimationcited by 76
  3. 2016Fast ε-free Inference of Simulation Models with Bayesian Conditional Density Estimationcited by 50
  4. 2022Investigating the Impact of Model Misspecification in Neural Simulation-based Inferencecited by 44
  5. 2019Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomycited by 36
  6. 2020Complete parameter inference for GW150914 using deep learningcited by 25
  7. 2016Inference Networks for Sequential Monte Carlo in Graphical Modelscited by 19
  8. 2020Gravitational-wave parameter estimation with autoregressive neural network flowscited by 19
  9. 2020Lightning-Fast Gravitational Wave Parameter Inference through Neural Amortizationcited by 17
  10. 2021Group equivariant neural posterior estimationcited by 16
  11. 2022alpha-Deep Probabilistic Inference (alpha-DPI): efficient uncertainty quantification from exoplanet astrometry to black hole feature extractioncited by 9
  12. 2021Swift sky localization of gravitational waves using deep learning seeded importance samplingcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 7; 2024: 7; 2025: 7; 2026: 6.

28 papers.

Sort:

By area:

  1. 2023Flow Matching for Scalable Simulation-Based Inferencecited by 97
  2. 2025Simulation-Based Inference: A Practical Guidecited by 30
  3. 2023Consistency Models for Scalable and Fast Simulation-Based Inferencecited by 22
  4. 2024Real-time gravitational-wave inference for binary neutron stars using machine learningcited by 21
  5. 2023Sensitivity-Aware Amortized Bayesian Inferencecited by 20
  6. 2023Adversarial robustness of amortized Bayesian inferencecited by 17
  7. 2024Diffusion posterior sampling for simulation-based inference in tall data settingscited by 17
  8. 2023Leveraging Self-Consistency for Data-Efficient Amortized Bayesian Inferencecited by 14
  9. 2025SciTS: Scientific Time Series Understanding and Generation with LLMscited by 13
  10. 2024Cost-aware simulation-based inferencecited by 6
  11. 2022Adapting to noise distribution shifts in flow-based gravitational-wave inferencecited by 5
  12. 2023Inferring Atmospheric Properties of Exoplanets with Flow Matching and Neural Importance Samplingcited by 4
16 more
  1. 2024Fast and Reliable Probabilistic Reflectometry Inversion with Prior-Amortized Neural Posterior Estimationcited by 4
  2. 2024Demonstrating the power and flexibility of variational assumptions for amortized neural posterior estimation in environmental applicationscited by 3
  3. 2024Amortized Bayesian Workflowcited by 3
  4. 2023nbi: the Astronomer's Package for Neural Posterior Estimationcited by 2
  5. 2026Conditional neural control variates for variance reduction in Bayesian inverse problemscited by 2
  6. 2024Accelerated Bayesian parameter estimation and model selection for gravitational waves with normalizing flowscited by 1
  7. 2025Synthesizing 3D Abstractions by Inverting Procedural Buildings with Transformerscited by 1
  8. 2025Discriminative versus Generative Approaches to Simulation-based Inferencecited by 1
  9. 2025Flexible Gravitational-Wave Parameter Estimation with Transformerscited by 1
  10. 2026Scalable Simulation-Based Model Inference with Test-Time Complexity Controlcited by 1
  11. 2026An Introduction to Bayesian and Frequentist Simulation-Based Inference with Machine Learningcited by 1
  12. 2025Recent Advances in Simulation-based Inference for Gravitational Wave Data Analysiscited by 0
  13. 2025Misspecification-robust amortised simulation-based inference using variational methodscited by 0
  14. 2026STEP: Scientific Time-Series Encoder Pretraining via Cross-Domain Distillationcited by 0
  15. 2026Neural Posterior Estimation for Stochastic Epidemic Models Using Final Outcome Datacited by 0
  16. 2026Fast, accurate, and differentiable: a neural-network surrogate for NRSur7dq4 precessing binary black hole waveformscited by 0