Abstract:Many areas of science make extensive use of computer simulators that implicitly encode intractable likelihood functions of complex systems. Classical statistical methods are poorly suited for these so-called likelihood-free inference (LFI) settings, especially outside asymptotic and low-dimensional regimes. At the same time, traditional LFI methods - such as Approximate Bayesian Computation or more recent machine learning techniques - do not guarantee confidence sets with nominal coverage in general settings (i.e., with high-dimensional data, finite sample sizes, and for any parameter value). In addition, there are no diagnostic tools to check the empirical coverage of confidence sets provided by such methods across the entire parameter space. In this work, we propose a unified and modular inference framework that bridges classical statistics and modern machine learning providing (i) a practical approach to the Neyman construction of confidence sets with frequentist finite-sample coverage for any value of the unknown parameters; and (ii) interpretable diagnostics that estimate the empirical coverage across the entire parameter space. We refer to the general framework as likelihood-free frequentist inference (LF2I). Any method that defines a test statistic can leverage LF2I to create valid confidence sets and diagnostics without costly Monte Carlo samples at fixed parameter settings. We study the power of two likelihood-based test statistics (ACORE and BFF) and demonstrate their empirical performance on high-dimensional, complex data. Code is available at <a class="link-external link-https" href="https://github.com/lee-group-cmu/lf2i" rel="external noopener nofollow">this https URL</a>.

What is the likelihood function, and how is it used in particle physics?

Empirical Likelihood Inference for Probability Density Functions under Association

The Simplified Likelihood Framework

Constructing model-agnostic likelihoods, a method for the reinterpretation of particle physics results

On the importance of factorization for fast binned likelihood inference

Learning likelihood ratios with neural network classifiers

Likelihood Equations and Scattering Amplitudes

A Law of Likelihood for Composite Hypotheses

Likelihood-free inference via classification

Approximate Differentiable Likelihoods for Astroparticle Physics Experiments

Boosting likelihood learning with event reweighting

Binned Likelihood including Monte Carlo Statistical Uncertainty in Bayesian Inference

Testable Likelihoods for Beyond-the-Standard Model Fits

PHYSTAT Informal Review: Marginalizing versus Profiling of Nuisance Parameters

On partial likelihood

Likelihood-Free Frequentist Inference: Bridging Classical Statistics and Machine Learning for Reliable Simulator-Based Inference

Generalised likelihood profiles for models with intractable likelihoods

Likelihood Correspondence of Statistical Models