Arbitrary Marginal Neural Ratio Estimation for Simulation-based Inference
arXiv:2110.00449
Abstract
In many areas of science, complex phenomena are modeled by stochastic parametric simulators, often featuring high-dimensional parameter spaces and intractable likelihoods. In this context, performing Bayesian inference can be challenging. In this work, we present a novel method that enables amortized inference over arbitrary subsets of the parameters, without resorting to numerical integration, which makes interpretation of the posterior more convenient. Our method is efficient and can be implemented with arbitrary neural network architectures. We demonstrate the applicability of the method on parameter inference of binary black hole systems from gravitational waves observations.
arXiv commentFourth Workshop on Machine Learning and the Physical Sciences (NeurIPS 2021)
References in corpus (13)
- Adam: A Method for Stochastic Optimization
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Observation of Gravitational Waves from a Binary Black Hole Merger
- Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)
- dynesty: A Dynamic Nested Sampling Package for Estimating Bayesian Posteriors and Evidences
- SGDR: Stochastic Gradient Descent with Warm Restarts
- Frequency-domain gravitational waves from non-precessing black-hole binaries. II. A phenomenological model for the advanced detector era
- Robust parameter estimation for compact binaries with ground-based gravitational-wave observations using the LALInference software library
- Dynamic nested sampling: an improved algorithm for parameter estimation and evidence calculation
- Approximating Likelihood Ratios with Calibrated Discriminative Classifiers
- Automatic Posterior Transformation for Likelihood-Free Inference
- Lightning-Fast Gravitational Wave Parameter Inference through Neural Amortization
- Learning about an exponential amount of conditional distributions