Improving the evaluation of samplers on multi-modal targets
arXiv:2504.08916
Abstract
Addressing multi-modality constitutes one of the major challenges of sampling. In this reflection paper, we advocate for a more systematic evaluation of samplers towards two sources of difficulty that are mode separation and dimension. For this, we propose a synthetic experimental setting that we illustrate on a selection of samplers, focusing on the challenging criterion of recovery of the mode relative importance. These evaluations are crucial to diagnose the potential of samplers to handle multi-modality and therefore to drive progress in the field.
Accepted at ICLR 2025 workshop "Frontiers in Probabilistic Inference: Learning meets Sampling"