Approximate Bayesian Computation with the Wasserstein Distance
From MaRDI portal
Publication:5234401
Abstract: A growing number of generative statistical models do not permit the numerical evaluation of their likelihood functions. Approximate Bayesian computation (ABC) has become a popular approach to overcome this issue, in which one simulates synthetic data sets given parameters and compares summaries of these data sets with the corresponding observed values. We propose to avoid the use of summaries and the ensuing loss of information by instead using the Wasserstein distance between the empirical distributions of the observed and synthetic data. This generalizes the well-known approach of using order statistics within ABC to arbitrary dimensions. We describe how recently developed approximations of the Wasserstein distance allow the method to scale to realistic data sizes, and propose a new distance based on the Hilbert space-filling curve. We provide a theoretical study of the proposed method, describing consistency as the threshold goes to zero while the observations are kept fixed, and concentration properties as the number of observations grows. Various extensions to time series data are discussed. The approach is illustrated on various examples, including univariate and multivariate g-and-k distributions, a toggle switch model from systems biology, a queueing model, and a L'evy-driven stochastic volatility model.
Recommendations
- On parameter estimation with the Wasserstein distance
- Measuring dependence in the Wasserstein distance for Bayesian nonparametric models
- Bayesian learning with Wasserstein barycenters
- A Wasserstein-Type Distance in the Space of Gaussian Mixture Models
- Scalable Bayes via barycenter in Wasserstein space
- Wasserstein distance on finite spaces: statistical inference and algorithms
- Density estimation of multivariate samples using Wasserstein distance
- Approximate Bayesian computation: a nonparametric perspective
- Projected Wasserstein Gradient Descent for High-Dimensional Bayesian Inference
- Approximation for the Wasserstein distance
Cited in
(47)- Dependence properties and Bayesian inference for asymmetric multivariate copulas
- Distance-learning for approximate Bayesian computation to model a volcanic eruption
- Summary statistics and discrepancy measures for approximate Bayesian computation via surrogate posteriors
- Limitations of the Wasserstein MDE for univariate data
- Wasserstein statistics in one-dimensional location-scale models
- Weighted approximate Bayesian computation via Sanov's theorem
- Minimax confidence intervals for the sliced Wasserstein distance
- A comparison of likelihood-free methods with and without summary statistics
- Likelihood-free inference with deep Gaussian processes
- Using space filling curves to compare two multivariate distributions with distribution-free tests
- Posterior asymptotics in Wasserstein metrics on the real line
- Approximate Bayesian computations to fit and compare insurance loss models
- Spectral density-based and measure-preserving ABC for partially observed diffusion processes. An illustration on Hamiltonian SDEs
- On parameter estimation with the Wasserstein distance
- Optimal Transport to a Variety
- scientific article; zbMATH DE number 7625190 (Why is no real title available?)
- Projected Wasserstein Gradient Descent for High-Dimensional Bayesian Inference
- GAT–GMM: Generative Adversarial Training for Gaussian Mixture Models
- Sequential Monte Carlo samplers to fit and compare insurance loss models
- Metropolis–Hastings via Classification
- Discrepancy-based inference for intractable generative models using quasi-Monte Carlo
- Unsupervised mixture estimation via approximate maximum likelihood based on the Cramér-von Mises distance
- On predictive inference for intractable models via approximate Bayesian computation
- Convergence rates for ansatz‐free data‐driven inference in physically constrained problems
- Approximating Bayes in the 21st century
- Black-box Bayesian inference for agent-based models
- Approximate Bayesian computation via classification
- Learning to simulate sequentially generated data via neural networks and Wasserstein training
- The distance between: an algorithmic approach to comparing stochastic models to time-series data
- Calibration of stochastic, agent-based neuron growth models with approximate Bayesian computation
- Generalized Bayesian likelihood-free inference
- Statistical inference with regularized optimal transport
- Box confidence depth: simulation-based inference with hyper-rectangles
- Bayesian nonparametric model-based clustering with intractable distributions: an ABC approach
- Approximate Bayesian inference in a model for self-generated gradient collective cell movement
- Inference for the stochastic FitzHugh-Nagumo model from real action potential data via approximate Bayesian computation
- On the Wasserstein Median of Probability Measures
- Composite goodness-of-fit tests with kernels
- Sliced-Wasserstein distances and flows on Cartan-Hadamard manifolds
- Sequential Monte Carlo ABC: an overview with application to COVID-19 data
- Concentration of discrepancy-based approximate Bayesian computation via Rademacher complexity
- Approximate Bayesian computation using the Fourier integral theorem
- Interpretable Model Summaries Using the Wasserstein Distance
- Parameter estimation for the fractional Hawkes process
- Robust probabilistic inference via a constrained transport metric (with discussion)
- Revisiting concentration results for approximate Bayesian computation
- Joint likelihood-free inference of the number of selected single nucleotide polymorphisms and their selection coefficients in an evolving population
This page was built for publication: Approximate Bayesian Computation with the Wasserstein Distance
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5234401)
