Nonparametric likelihood-free inference with Jensen-Shannon divergence for simulator-based models with categorical output
This work addresses the need for efficient frequentist inference in applications with big data, providing a computationally efficient method for simulator-based models, though it is incremental as it builds on existing divergence-based approaches.
The paper tackles the problem of frequentist inference for simulator-based models with categorical output, which has received less attention compared to Bayesian methods, by deriving theoretical results that enable estimation, hypothesis testing, and confidence interval construction using asymptotic properties of the Jensen-Shannon divergence, offering a rapid alternative to computation-intensive approaches.
Likelihood-free inference for simulator-based statistical models has recently attracted a surge of interest, both in the machine learning and statistics communities. The primary focus of these research fields has been to approximate the posterior distribution of model parameters, either by various types of Monte Carlo sampling algorithms or deep neural network -based surrogate models. Frequentist inference for simulator-based models has been given much less attention to date, despite that it would be particularly amenable to applications with big data where implicit asymptotic approximation of the likelihood is expected to be accurate and can leverage computationally efficient strategies. Here we derive a set of theoretical results to enable estimation, hypothesis testing and construction of confidence intervals for model parameters using asymptotic properties of the Jensen--Shannon divergence. Such asymptotic approximation offers a rapid alternative to more computation-intensive approaches and can be attractive for diverse applications of simulator-based models. 61