Likelihood-free inference via classification
From MaRDI portal
Abstract: Increasingly complex generative models are being used across disciplines as they allow for realistic characterization of data, but a common difficulty with them is the prohibitively large computational cost to evaluate the likelihood function and thus to perform likelihood-based statistical inference. A likelihood-free inference framework has emerged where the parameters are identified by finding values that yield simulated data resembling the observed data. While widely applicable, a major difficulty in this framework is how to measure the discrepancy between the simulated and observed data. Transforming the original problem into a problem of classifying the data into simulated versus observed, we find that classification accuracy can be used to assess the discrepancy. The complete arsenal of classification methods becomes thereby available for inference of intractable generative models. We validate our approach using theory and simulations for both point estimation and Bayesian inference, and demonstrate its use on real data by inferring an individual-based epidemiological model for bacterial infections in child care centers.
Recommendations
Cites work
- A Method of Simulated Moments for Estimation of Discrete Response Models Without Numerical Integration
- A tutorial on Bayesian models of perception
- A user's guide to measure theoretic probability
- Approximate Bayesian computational methods
- Asymptotic Statistics
- Bayesian optimization for likelihood-free inference of simulator-based statistical models
- Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation. With discussion and authors' reply
- Encompassing and indirect inference
- Estimating the transmission dynamics of \textit{Streptococcus pneumoniae} from strain prevalence data
- scientific article; zbMATH DE number 2038320 (Why is no real title available?)
- scientific article; zbMATH DE number 3894273 (Why is no real title available?)
- scientific article; zbMATH DE number 6276120 (Why is no real title available?)
- scientific article; zbMATH DE number 2238781 (Why is no real title available?)
- Likelihood-free inference via classification
- Simulation and the Asymptotics of Optimization Estimators
- Statistical behavior and consistency of classification methods based on convex risk minimization.
- The Poisson transform for unnormalised statistical models
- Weak convergence and empirical processes. With applications to statistics
Cited in
(36)- A likelihood-MPEC approach to target classification
- Inverse reinforcement learning from summary data
- Asymptotically exact inference in differentiable generative models
- Likelihood-free inference via classification
- Variational Bayes with synthetic likelihood
- Efficient acquisition rules for model-based approximate Bayesian computation
- Likelihood-free inference in high dimensions with synthetic likelihood
- Distance-learning for approximate Bayesian computation to model a volcanic eruption
- Multimodal image synthesis with conditional implicit maximum likelihood estimation
- GADE: a generative adversarial approach to density estimation and its applications
- Parallel Gaussian process surrogate Bayesian inference with noisy likelihood evaluations
- Summary statistics and discrepancy measures for approximate Bayesian computation via surrogate posteriors
- Minimax confidence intervals for the sliced Wasserstein distance
- A comparison of likelihood-free methods with and without summary statistics
- New approximate Bayesian computation algorithm for censored data
- Local dimension reduction of summary statistics for likelihood-free inference
- Bayesian model-based clustering for longitudinal ordinal data
- Bayesian optimization for likelihood-free inference of simulator-based statistical models
- Bayesian inference of spreading processes on networks
- scientific article; zbMATH DE number 7625191 (Why is no real title available?)
- The frontier of simulation-based inference
- Towards end‐to‐end likelihood‐free inference with convolutional neural networks
- scientific article; zbMATH DE number 7306864 (Why is no real title available?)
- Mining gold from implicit models to improve likelihood-free inference
- Classification by likelihood accordance functions
- ABC of the future
- Likelihood-free inference by ratio estimation
- Metropolis–Hastings via Classification
- Discrepancy-based inference for intractable generative models using quasi-Monte Carlo
- Sequential Bayesian experimental design for implicit models via mutual information
- Approximate Bayesian computation via classification
- Approximate Bayesian computation for the natural history of breast cancer, with application to data from a Milan cohort study
- Concentration of discrepancy-based approximate Bayesian computation via Rademacher complexity
- Approximate Bayesian inference based on expected evaluation
- Leveraging nested MLMC for sequential neural posterior estimation with intractable likelihoods
- Robust Bayesian synthetic likelihood via a semi-parametric approach
This page was built for publication: Likelihood-free inference via classification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1702017)
