Papers

Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models

J Vetter, M Gloeckler, D Gedon, JH Macke - arXiv preprint arXiv …, 2025 - arxiv.org
Computer Science paper cs.LG Suggest

… 2.2 TabPFN for simulation-based inference To perform simulation-based inference with TabPFN, we repurpose it as a general conditional density estimator. The key to …

Cited by Link to paper

BibTeX

@article{2504.17660v1,
Author        = {Julius Vetter and Manuel Gloeckler and Daniel Gedon and Jakob H. Macke},
Title         = {Effortless, Simulation-Efficient Bayesian Inference using Tabular
  Foundation Models},
Eprint        = {2504.17660v1},
ArchivePrefix = {arXiv},
PrimaryClass  = {cs.LG},
Abstract      = {Simulation-based inference (SBI) offers a flexible and general approach to
performing Bayesian inference: In SBI, a neural network is trained on synthetic
data simulated from a model and used to rapidly infer posterior distributions
for observed data. A key goal for SBI is to achieve accurate inference with as
few simulations as possible, especially for expensive simulators. In this work,
we address this challenge by repurposing recent probabilistic foundation models
for tabular data: We show how tabular foundation models -- specifically TabPFN
-- can be used as pre-trained autoregressive conditional density estimators for
SBI. We propose Neural Posterior Estimation with Prior-data Fitted Networks
(NPE-PF) and show that it is competitive with current SBI approaches in terms
of accuracy for both benchmark tasks and two complex scientific inverse
problems. Crucially, it often substantially outperforms them in terms of
simulation efficiency, sometimes requiring orders of magnitude fewer
simulations. NPE-PF eliminates the need for inference network selection,
training, and hyperparameter tuning. We also show that it exhibits superior
robustness to model misspecification and can be scaled to simulation budgets
that exceed the context size limit of TabPFN. NPE-PF provides a new direction
for SBI, where training-free, general-purpose inference models offer efficient,
easy-to-use, and flexible solutions for a wide range of stochastic inverse
problems.},
Year          = {2025},
Month         = {Apr},
Url           = {http://arxiv.org/abs/2504.17660v1},
File          = {2504.17660v1.pdf}
}

Share