Guiding Posterior Exploration with Optimizer-Derived Geometry

July 28, 2026 ยท Grace Period ยท ๐Ÿ› AISTATS 2026

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Moritz Schlager, Emanuel Sommer, Thomas Mรถllenhoff, David Rรผgamer arXiv ID 2607.25312 Category cs.LG: Machine Learning Citations 0 Venue AISTATS 2026
Abstract
Sampling-based methods offer a principled approach to uncertainty quantification in Bayesian neural networks. Their practical use, however, is often challenged by the computational cost of exploring high-dimensional and multimodal posterior distributions. To overcome these difficulties, Bayesian Deep Ensembles, i.e., warmstarting the sampling from several optimized solutions, have proven to be an effective strategy. In this paper, we demonstrate that curvature estimates computed during the warmstart as a byproduct in adaptive optimizers such as AdamW can inform the sampling phase at negligible additional cost. Specifically, our proposed preconditioned sampling strategy based on optimizer-derived geometries can substantially reduce or even eliminate the need for a lengthy sampling burn-in phase and leads to greater numerical stability. This approach consistently maintains or improves predictive performance and uncertainty quantification without any additional computational costs. We confirm the consistency of our findings across various datasets and network architectures.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning