En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)

PhD contract (M/W) - AI and archaeozoology: learning methods for the identification and clustering of animal remains

This offer is available in the following languages:
- Français-- Anglais

Date Limite Candidature : lundi 9 octobre 2023

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

Informations générales

Intitulé de l'offre : PhD contract (M/W) - AI and archaeozoology: learning methods for the identification and clustering of animal remains (H/F)
Référence : UMR7264-ANNGOM-004
Nombre de Postes : 1
Lieu de travail : NICE
Date de publication : lundi 18 septembre 2023
Type de contrat : CDD Doctorant/Contrat doctoral
Durée du contrat : 36 mois
Date de début de la thèse : 15 octobre 2023
Quotité de travail : Temps complet
Rémunération : 2 135,00 € gross monthly
Section(s) CN : Interactions, particles, nuclei, from laboratory to cosmos

Description du sujet de thèse

The inspection of faunal remains found in archaeological contexts provides archaeozoological researchers with information about past relationships between humans and non-human mammals, paleoenvironments, past animal populations (biology) and the subsistence economy of ancient societies.These remains are mainly complete and/or incomplete bones that have been altered by man and the environment over a long period.The process of anatomical and taxonomic identification from the bones is carried out by archaeozoologists mainly on the basis of anatomical, morphological and biometric criteria.This is long, tedious and particularly demanding work, and the development of original AI tools adapted to assist experts in taxonomic identification is of particular interest.The thesis will therefore focus on two areas: i) interactive supervised learning for the identification of morphologically similar herbivore species and ii) unsupervised learning for the study of the morphological evolution of goat bones. Axis i) will focus on the development of hybrid neural architectures of the PointSIFT type whose input layers could be represented by anatomical descriptors provided by the expert. Other learning approaches based on topological data analysis (TDA) will also be considered, as well as symbolic AI techniques for integrating expert knowledge into machine learning models. Axis ii) will be at the intersection between geometric morphometry and machine learning and will focus mainly on the development of clustering and dictionary learning techniques based on optimal transport (typically Procrustes-Wasserstein analysis).

Contexte de travail

The candidate will have a master's degree in mathematics, computer science or physics and an in-depth knowledge of statistical learning (machine learning). Knowledge or previous experience in archaeology and archaeozoology will be a strong asset. He/she will adopt a multidisciplinary approach (archaeozoology, geometric morphometry, mathematics and computer science) and will be required to code in Python/R. He/she will be based at the INRIA Université Côte d'Azur centre in Sophia Antipolis and at the CEPAM-CNRS laboratory in Nice. The doctoral contract is part of Mr Corneli's Junior Professorship and the thesis will be labelled 3IA (Institut 3IA Côte d'Azur label). In practical terms, the PhD student will be subject to the teaching (64 hours per year for 3 years) and training obligations laid down for students funded by the Institut 3IA. The candidate will be supervised by Emmanuelle Vila, an archaeozoologist specialising in animal exploitation in the Ancient Near East and Eastern Mediterranean, co-supervised by Marco Corneli, an expert in statistical learning, and co-supervised by Manon Vuillien, an archaeozoologist.