The Library
Robust Bayesian inference for simulator-based models via the MMD posterior bootstrap
Tools
Tools
Tools
Dellaporta, Charita, Knoblauch, Jeremias, Damoulas, Theodoros and Briol, François-Xavier (2022) Robust Bayesian inference for simulator-based models via the MMD posterior bootstrap. In: 25th International Conference in Artificial Intelligence and Statistics (AISTATS), Virtual, 28-30 Mar 2022. Published in: Proceedings of Machine Learning Research, 151 ISSN 2640-3498. (In Press)
|
PDF
WRAP-robust-Bayesian-inference-simulator-based-models-via-the-MMD-posterior-bootstrap-Dellaporta-2022.pdf - Accepted Version - Requires a PDF viewer. Download (1263Kb) | Preview |
Abstract
This paper responds to recent calls to further incorporate the study of animal healthcare into the sociology of health and illness. It focuses on a theme with a long tradition in medical sociology, namely clinical communication, but explores matters distinctive to veterinary practice. Drawing on video recordings of 60 consultations across three small animal veterinary clinics in the UK, we explore how clients and veterinarians (or ‘vets’) fashion fleeting “coalitions of touch”, that aptly position the animal to enable the performance of medical work, often in the face of physical resistance. Building on recent developments in the study of haptic sociality, we analyse how care and emotional concern for animal patients is communicated through various forms of embodied action; thus, how the problematics of forced care and restraint are mitigated through distinctive ways of touching and holding animal patients. Moreover, while prior studies of small animal veterinary work have highlighted the significance of talk within the clinician-animal-client triad, we reveal the fundamentally embodied and collaborative work of managing and controlling patients during sometimes intense and fast-moving episodes of veterinary care. Simulator-based models are models for which the likelihood is intractable but simulation of synthetic data is possible. They are often used to describe complex real-world phenomena, and as such can often be misspecified in practice. Unfortunately, existing Bayesian approaches for simulators are known to perform poorly in those cases. In this paper, we propose a novel algorithm based on the posterior bootstrap and maximum mean discrepancy estimators. This leads to a highly-parallelisable Bayesian inference algorithm with strong robustness properties. This is demonstrated through an in-depth theoretical study which includes generalisation bounds and proofs of frequentist consistency and robustness of our posterior. The approach is then assessed on a range of examples including a g-and-k distribution and a toggle-switch model.
| Item Type: | Conference Item (Paper) | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Subjects: | Q Science > QA Mathematics | ||||||||||||||||||
| Divisions: | Faculty of Science, Engineering and Medicine > Science > Computer Science Faculty of Science, Engineering and Medicine > Science > Statistics |
||||||||||||||||||
| Library of Congress Subject Headings (LCSH): | Bayesian statistical decision theory, Bootstrap (Statistics), Machine learning -- Statistical methods, Bootstrap (Computer program) | ||||||||||||||||||
| Journal or Publication Title: | Proceedings of Machine Learning Research | ||||||||||||||||||
| Publisher: | ML Research Press | ||||||||||||||||||
| ISSN: | 2640-3498 | ||||||||||||||||||
| Official Date: | 2022 | ||||||||||||||||||
| Dates: |
|
||||||||||||||||||
| Volume: | 151 | ||||||||||||||||||
| Status: | Peer Reviewed | ||||||||||||||||||
| Publication Status: | In Press | ||||||||||||||||||
| Access rights to Published version: | Restricted or Subscription Access | ||||||||||||||||||
| Date of first compliant deposit: | 15 February 2022 | ||||||||||||||||||
| Date of first compliant Open Access: | 18 February 2022 | ||||||||||||||||||
| RIOXX Funder/Project Grant: |
|
||||||||||||||||||
| Conference Paper Type: | Paper | ||||||||||||||||||
| Title of Event: | 25th International Conference in Artificial Intelligence and Statistics (AISTATS) | ||||||||||||||||||
| Type of Event: | Conference | ||||||||||||||||||
| Location of Event: | Virtual | ||||||||||||||||||
| Date(s) of Event: | 28-30 Mar 2022 | ||||||||||||||||||
| Related URLs: | |||||||||||||||||||
| Open Access Version: |
Request changes or add full text files to a record
Repository staff actions (login required)
 |
View Item |
Downloads
Downloads per month over past year
