Dear all,

the next OWABI seminar www.warwick.ac.uk/owabi of the Season is quickly approaching!

We are glad to inform you that our next speaker is Kate Lee(University of Auckland), who will talk about "Towards Robust and Scalable Bayesian Learning" on the 29th May at 10am UK time (note the different time!), with an abstract reported below.

The talk will be streamed on the OWABI MS Teams Channel General | OWABI Seminar: One World Approximate Bayesian Inference Seminar at this link 

https://teams.microsoft.com/meet/382746871856192?p=U7SJQ144q6hLNVFGdS
Meeting ID: 382 746 871 856 192
Passcode: FZ2gn73K

SpeakerKate Lee (University of Auckland)
Title: Towards Robust and Scalable Bayesian Learning
Abstract: My talk consists of two parts. The first focuses on Generalised Bayesian Inference (GBI), where the loss hyperparameters are critical under model misspecification yet difficult to determine in a principled way. We introduce a Bayesian framework for hyperparameter inference using held-out data, enabling coherent estimation and uncertainty quantification, together with amortized generalized-posterior approximations that avoid repeated costly sampling across datasets and hyperparameter values. The second addresses scalability in simulation-based inference through data reduction, developing methods that learn informative low-dimensional summaries to enable efficient inference in complex settings, with applications to gravitational wave analysis.
Keywords: GBI, SBI, amortized learning

We're looking forward to seeing you next week,
best,
Massimiliano on the behalf of the OWABI Seminar Organisers


------
Dr. Massimiliano Tamborrino SFHEA
Reader (Associate Professor)
Department of Statistics
WIHEA Fellow and Internationalisation Learning Circle Co-Lead
University of Warwick
https://warwick.ac.uk/tamborrino