PolySwyft: sequential simulation-based nested sampling

2025/12/09 by Scheutwinkel, Kilian H., Handley, Will, Weniger, Christoph +1
Computer Science · Physics and Astronomy · #Cosmology and Nongalactic Astrophysics (astro-ph.CO) #FOS: Physical sciences #Galaxies: Formation, Evolution, Phenomena #Gaussian Processes and Bayesian Inference #Instrumentation and Methods for Astrophysics (astro-ph.IM) #Statistical Mechanics and Entropy

paper · doi:10.48550/arxiv.2512.08316

openalex publication_date 2025/12/09 · openalex created_date 2025/12/11 · openalex updated_date 2026/07/28

Abstract

We present PolySwyft, a novel, non-amortised simulation-based inference framework that unites the strengths of nested sampling (NS) and neural ratio estimation (NRE) to tackle challenging posterior distributions when the likelihood is intractable but a forward simulator is available. By nesting rounds of NRE within the exploration of NS, and employing a principled KL-divergence criterion to adaptively terminate sampling, PolySwyft achieves faster convergence on complex, multimodal targets while rigorously preserving Bayesian validity. On a suite of toy problems with analytically known posteriors of a dim(theta,D)=(5,100) multivariate Gaussian and multivariate correlated Gaussian mixture model, we demonstrate that PolySwyft recovers all modes and credible regions with fewer simulator calls than swyft's TNRE. As a real-world application, we infer cosmological parameters dim(theta,D)=(6,111) from CMB power spectra using CosmoPower. PolySwyft is released as open-source software, offering a flexible toolkit for efficient, accurate inference across the astrophysical sciences and beyond.

Citations

• Dark Energy Survey Year 3 results: likelihood-free, simulation-based wCDM inference with neural compression of weak-lensing map statistics
• aeons: approximating the end of nested sampling
• Simulation-based Inference for High-dimensional Data using Surjective Sequential Neural Likelihood Estimation
• Sequential Neural Posterior and Likelihood Approximation
• ArXiv
• Flexible statistical inference for mechanistic models of neural dynamics
• Likelihood-free inference by ratio estimation
• Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
• Adam: A Method for Stochastic Optimization

Related

• Simulation-efficient marginal posterior estimation with swyft: stop wasting your precious time
• SuperNest: accelerated nested sampling applied to astrophysics and cosmology
• High-Dimensional Bayesian Model Comparison in Cosmology with GPU-accelerated Nested Sampling and Neural Emulators
• Cosmological Parameter Estimation with Sequential Linear Simulation-based Inference
• Nested sampling for physical scientists