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.
@article{arxiv.2504.08916,
title = {Improving the evaluation of samplers on multi-modal targets},
author = {Louis Grenioux and Maxence Noble and Marylou Gabrié},
journal= {arXiv preprint arXiv:2504.08916},
year = {2025}
}
Comments
Accepted at ICLR 2025 workshop "Frontiers in Probabilistic Inference: Learning meets Sampling"