BibTeX

@article{2510.13763v1,
Author        = {Yang Yang and Severi Rissanen and Paul E. Chang and Nasrulloh Loka and Daolang Huang and Arno Solin and Markus Heinonen and Luigi Acerbi},
Title         = {PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inference},
Eprint        = {2510.13763v1},
ArchivePrefix = {arXiv},
PrimaryClass  = {stat.ML},
Abstract      = {Amortized simulator-based inference offers a powerful framework for tackling
Bayesian inference in computational fields such as engineering or neuroscience,
increasingly leveraging modern generative methods like diffusion models to map
observed data to model parameters or future predictions. These approaches yield
posterior or posterior-predictive samples for new datasets without requiring
further simulator calls after training on simulated parameter-data pairs.
However, their applicability is often limited by the prior distribution(s) used
to generate model parameters during this training phase. To overcome this
constraint, we introduce PriorGuide, a technique specifically designed for
diffusion-based amortized inference methods. PriorGuide leverages a novel
guidance approximation that enables flexible adaptation of the trained
diffusion model to new priors at test time, crucially without costly
retraining. This allows users to readily incorporate updated information or
expert knowledge post-training, enhancing the versatility of pre-trained
inference models.},
Year          = {2025},
Month         = {Oct},
Url           = {http://arxiv.org/abs/2510.13763v1},
File          = {2510.13763v1.pdf}
}