Séminaire de Probabilités et Statistique
Monday 13 April 2026 à 13:45 - UM, campus Triolet, bâtiment 9, salle 109 (1er étage)
Marco Corneli (Université Côté d'Azur)
Archaeozoology and machine learning: a promising matching
The talk will focus on two case studies from archaeozoology we dealt with in the last few years in the context of my chair of PJ in AI for history and archaeology. The former is a supervised learning problem, aiming to identify animal species from bone remains. By using descriptors from Topological Data Analysis, we revisited multiple kernel learning under the lenses of recent improvements in importance weighted variational inference. The latter investigates some properties of an optimal transport distance (Procrustes-Wasserstein), encapsulating invariance with respect to rigid transformations, in order to simulate the evolution in time of a sheep's astragalus from the Calcholitic to modern era.
Séminaire en salle 109, également retransmis sur zoom : https://umontpellier-fr.zoom.us/j/7156708132