The 12th edition of PGM took place from September 11th to 13th, 2024, in Nijmegen, in the Lindenberg venue. PGM 2024 (www.ru.nl/pgm2024) was attended by 71 participants from 18 countries. 40 papers were submitted of which 32 were accepted for presentation and publication in the Proceedings of Machine Learning Research (https://proceedings.mlr.
A broad range of topics within the PGM sub-discipline of AI was covered, ranging from causality to decision making, from continuous models to probabilistic circuits, and from inference to learning. In addition to the regular sessions there were two fascinating keynote lectures: one by Arto Klami on the choice of prior distributions, and one by Sara Magliacane on causal learning in temporal settings. The scientific part of the conference was interspersed by social activities: a city tour of Nijmegen and Reception in De Waagh on Wednesday 11th and the conference dinner in De Hemel on Thursday 12th. At the closing Community Meeting on Friday, the Best Student Paper award was offered to Anna Rodum Bjøru for the paper “A Divide and Conquer Approach for Solving Structural Causal Models” co-authored by Anna Rodum Bjøru, Rafael Cabañas de Paz, Helge Langseth, and Antonio Salmeron.
At this meeting, also the venue for PGM 2026 was elected: we will return to Spain as Valencia will host the conference in 2026. Prior to the conference, the PGM group at Radboud University and the Personalised Care in Oncology programme hosted a co-located informal workshop on PGMs in Healthcare, focusing on the development, CE certification, and deployment of Bayesian networks as clinical decision support systems in healthcare.
