Valeria Vitelli (University of Oslo)
will speak on
An overview over Bayesian rank-based clustering of high-dimensional data using the Mallows rank model
Time: 3:00PM
Date: Thu 13th November 2025
Location: E0.32 (beside Pi restaurant)
[map]
Abstract: The Mallows model is a popular model for rankings, used to estimate individual behaviours and preferences in several areas, such as marketing and politics. It flexibly adapts to different types of preference data, and the previously proposed Bayesian Mallows Model (BMM) offers a computationally efficient framework for Bayesian inference. However, scaling is poor and the model becomes less realistic when the pool of items is large or ultra-large. In this talk, I will introduce recent extensions of BMM in a number of directions: (i) embedding covariate information related to assessors; (ii) combining a lower-dimensional version of BMM (lowBMM) more suited to modeling large datasets with a Bayesian mixture of Mallows models; (iii) finally, devising a variational inferential Pseudo-Mallows approach.
(This talk is part of the Statistics and Actuarial Science series.)
PDF notice
Return to all seminars