PhD (M/F) Differentiable hyperbolicity and Graph Neural Networks

Descrizione

PhD (M/F) Differentiable hyperbolicity and Graph Neural Networks Sintesi in italiano (traduzione automatica): Il candidato sarà inserito nel team Compact di IRISA, uno dei più grandi laboratori di ricerca in Francia nel campo dell'informatica e della tecnologia dell'informazione. La posizione si concentra sull'integrazione e la valutazione di un nuovo concetto di iperbolicità differenziabile nelle reti neurali grafiche. Il laboratorio si occupa di vari ambiti, tra cui intelligenza artificiale, sicurezza informatica e analisi dei Big Data. È richiesta una laurea in informatica o un campo correlato. La posizione è soggetta all'autorizzazione dell'autorità competente secondo le normative vigenti. Il candidato dovrà avere competenze in geometria riemanniana e machine learning applicato ai grafi. The candidate will be attached to IRISA's Compact team. About the lab ============= www.irisa.fr IRISA is one of France's largest research laboratories (over 850 people) in the field of computer science and information technology. Structured into seven scientific departments, IRISA is a laboratory of excellence whose scientific priorities are bioinformatics, systems security, new software architectures, virtual reality, Big Data analysis and artificial intelligence. IRISA is at the heart of the digital transition of society and innovation in the fields of cybersecurity, health, environment and ecology, transport, robotics, energy, culture and artificial intelligence. The position is located in a sector covered by the protection of scientific and technical potential (PPST), and therefore requires, in accordance with regulations, that your arrival be authorized by the competent MESR authority. Several concepts of Riemannian geometry have recently found applications in graph Machine Learning, such as the Ollivier-Ricci curvature. A recent example is the delta-hyperbolicity, which measures how graphs locally look like a tree, which has found applications in treating over-squashing in Graph Neural Networks. However, a fundamental limit of this notion is its combinatorial nature, which limits its integration in ML pipelines. Recently, a differentiable surrogate of delta-hyperbolicity has been proposed, and studied theoretically. The goal of this project is to integrate and benchmark this new notion in Graph ML and Graph Neural Networks. Annuncio in inglese. Fonte: Euraxess (Commissione europea).