PreprintCC BY 4.0arXiv
Bayesian inference under model misspecification
Abstract
The likelihood input to a Bayesian analysis almost never exactly represents how the data were generated, calling into question the validity of posterior inferences. We review a variational interpretation of the Bayesian posterior as an alternative justification for its use under model misspecification, and consider the resulting implications on uncertainty quantification and parameter estimation. We then introduce a range of techniques that seek to obtain generalised Bayesian inferences that account for model misspecification
Versions
Related
- Overcoming Model Misspecification in Bayesian Inference of Molecular Signalling Networks
- When Noise Estimation Hides Basis Misspecification in Repeated Bayesian Inverse Problems
- Scalable Diffusion SBI for Compositional Inference under Simulator Misspecification
- Efficient Bayesian inference for multiple network data
- Modern Bayesian Modeling and Inference
BibTeX
@misc{jewson2026bayesian,
title = {{Bayesian inference under model misspecification}},
author = {Jack Jewson},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11173},
eprint = {2610.11173},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11173},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Bayesian inference under model misspecification
AU - Jack Jewson
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.11173
UR - https://arxiv.org/abs/2610.11173
AB - The likelihood input to a Bayesian analysis almost never exactly represents how the data were generated, calling into question the validity of posterior inferences. We review a variational interpretation of the Bayesian posterior as an alternative justification for its use under model misspecification, and consider the resulting implications on uncertainty quantification and parameter estimation. We then introduce a range of techniques that seek to obtain generalised Bayesian inferences that account for model misspecification
ER -