Improving the evaluation of samplers on multi-modal targets
This work addresses the problem of systematic evaluation for samplers in multi-modal settings, which is crucial for diagnosing sampler potential and driving progress in the field, but it is incremental as it builds on existing evaluation methods.
The paper tackles the challenge of evaluating samplers on multi-modal targets by proposing a synthetic experimental setting to assess mode separation and dimension, focusing on recovery of mode relative importance, and illustrates this on a selection of samplers.
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.