Simulation-efficient marginal posterior estimation with swyft: stop wasting your precious time

11/27/2020
by   Benjamin Kurt Miller, et al.
9

We present algorithms (a) for nested neural likelihood-to-evidence ratio estimation, and (b) for simulation reuse via an inhomogeneous Poisson point process cache of parameters and corresponding simulations. Together, these algorithms enable automatic and extremely simulator efficient estimation of marginal and joint posteriors. The algorithms are applicable to a wide range of physics and astronomy problems and typically offer an order of magnitude better simulator efficiency than traditional likelihood-based sampling methods. Our approach is an example of likelihood-free inference, thus it is also applicable to simulators which do not offer a tractable likelihood function. Simulator runs are never rejected and can be automatically reused in future analysis. As functional prototype implementation we provide the open-source software package swyft.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/02/2021

Truncated Marginal Neural Ratio Estimation

Parametric stochastic simulators are ubiquitous in science, often featur...
research
11/15/2021

Fast and Credible Likelihood-Free Cosmology with Truncated Marginal Neural Ratio Estimation

Sampling-based inference techniques are central to modern cosmological d...
research
03/08/2022

Variational methods for simulation-based inference

We present Sequential Neural Variational Inference (SNVI), an approach t...
research
02/01/2023

Computational Discovery of Microstructured Composites with Optimal Strength-Toughness Trade-Offs

The conflict between strength and toughness is a fundamental problem in ...
research
09/04/2023

perms: Marginal likelihood estimation for binary Bayesian nonparametric models in Python and R

Binary responses arise in a multitude of statistical problems, including...
research
02/10/2020

On Contrastive Learning for Likelihood-free Inference

Likelihood-free methods perform parameter inference in stochastic simula...
research
06/06/2015

Approximating Likelihood Ratios with Calibrated Discriminative Classifiers

In many fields of science, generalized likelihood ratio tests are establ...

Please sign up or login with your details

Forgot password? Click here to reset