Sampling via Gaussian Mixture Approximations
This provides a computationally efficient method for sampling in Bayesian inference and machine learning, though it is incremental as it builds on existing mixture and resampling techniques.
The paper tackles the problem of sampling from unnormalized target densities by introducing Gaussian Mixture Approximation (GMA) samplers, which use a two-stage gradient-free approach to efficiently produce accurate samples, as validated with empirical results showing improved speed and accuracy across diverse densities.
We present a family of \textit{Gaussian Mixture Approximation} (GMA) samplers for sampling unnormalised target densities, encompassing \textit{weights-only GMA} (W-GMA), \textit{Laplace Mixture Approximation} (LMA), \textit{expectation-maximization GMA} (EM-GMA), and further variants. GMA adopts a simple two-stage paradigm: (i) initialise a finite set of Gaussian components and draw samples from a proposal mixture; (ii) fit the mixture to the target by optimising either only the component weights or also the means and variances, via a sample-based KL divergence objective that requires only evaluations of the unnormalised density, followed by stratified resampling. The method is gradient-free, and computationally efficient: it leverages the ease of sampling from Gaussians, efficient optimisation methods (projected gradient descent, mirror descent, and EM), and the robustness of stratified resampling to produce samples faithful to the target. We show that this optimisation-resampling scheme yields consistent approximations under mild conditions, and we validate this methodology with empirical results demonstrating accuracy and speed across diverse densities.