Economics and Computation Seminar Series http://www.cs.ox.ac.uk/seminars/

This interdisciplinary seminar series covers topics at the intersection of computer science, economics, and mathematics - in particular, computational social choice and algorithmic game theory. Join us for exciting talks and engaging discussions! Seminars usually take place on Fridays at 2pm, unless stated otherwise. For additional reminders, please contact Nicholas to be added to the seminar mailing list.

 en-gb 360 Thu, 15 Oct 2026 11:39:26 GMT Thu, 15 Oct 2026 11:39:26 GMT Economics and Computation Seminar Series http://blogs.law.harvard.edu/tech/rss/ Proportional Hierarchical Clustering. Leon Kellerhals (TU Clausthal) http://www.cs.ox.ac.uk/seminars/2859.html <p>Hierarchical clustering represents data at multiple levels of granularity, but proportional fairness is traditionally only defined for a single partition. We therefore study how it can be extended to an entire hierarchy. Building on prior work on proportional clustering, we introduce two notions: the local core, which requires proportional fairness at each split, and the global core, which requires it at every partition in the hierarchy. We study both notions under the assumption that an agent's loss (disutility) for a clustering is the maximum distance to other members of their cluster. We show that the two core notions are incomparable and that standard agglomerative clustering techniques fail to satisfy them. However, under the maximum loss objective, a hierarchical clustering in the local core always exists. On the way, we resolve an open question from Caragiannis et al. (2024) by showing that there always exists a non-centroid clustering in the core for k = 2, whereas for k &ge; 3 the core can be empty (even for Euclidean distances). For the global core, we show that even arbitrary approximations of it need not exist. However, doubling the coalition size threshold allows us to compute an outcome in the global 16-core for maximum-loss and global 8n-core for average-loss, where n is the number of data points.<br />This is joint work with Robert Bredereck, Eva Deltl, and Jannik Peters.</p> <br/>Room 441 Economics and Computation Seminar Series oucl-seminar-Economics and Computation Seminar Series-2859 Thu, 15 Oct 2026 11:39:26 GMT