Dear all,

We are delighted to announce that our first speaker of the season is Chris Oates (Newcastle University) who will be talking about "Overcoming Model Misspecification in Bayesian Inference of Molecular Signalling Networks" on Thursday the 29^th October at 11am UK time with an abstract reported below.

The talk will be streamed at the OWABI MS Team channel General | OWABI Seminar: One World Approximate Bayesian Inference Seminar | Microsoft Teams at the following link

Join: https://teams.microsoft.com/meet/3744683199415937?p=eu2tG5Ukrf84zJvPmB

Speaker: Chris Oates (Newcastle University)

Title: Overcoming Model Misspecification in Bayesian Inference of Molecular Signalling Networks

Abstract: Bayesian inference of molecular signalling networks usually relies on tractability of the marginal likelihood, enabling the set of possible networks to be efficiently explored. As such, linear models with independent errors and conjugate priors are routinely used. However, the dynamics of molecular signalling are nonlinear, and relevant confounders are often unobserved; failure to account for these complexities will almost certainly lead to over-confident inferences in the standard Bayesian framework. To confront this reality, we develop a post-Bayesian approach to inference of molecular signalling networks, guided by the principle that uncertainty should not vanish when the statistical model is misspecified, even in the infinite-data limit. Technically, we extend the predictively-oriented (PrO) posterior of McLatchie et al. (2025) to the setting of latent variable models, empirically investigating the properties of PrO posteriors in the challenging network inference context.  This is joint work with Yinghan Li, Heishiro Kanagawa, Francesca Romana Crucinio, and Matthew A. Fisher.

Keywords: model misspecification, post-Bayesian, predictively-oriented posteriors

Reference: Y. Li, H. Kanagawa, F. R. Crucinio, M. A. Fisher, C.J. Oates. Overcoming Model Misspecification in Bayesian Inference of Molecular Signalling Networks. Preprint at ArXiv:2609.24590 https://arxiv.org/abs/2609.24590