The EU Rail Knowledge Graph
17 février 2025 @ 10 h 00 min – 12 h 00 min –
Salle 104 du bâtiment 11 dit le château, 2 Place Pierre Viala Campus La Gaillarde, 34000 Montpellier
SESAME – SEmantic web SeminAr MontpEllier
The European Union Agency for Railways (ERA) is an Agency in charge of facilitating the implementation of an efficient, safe, and interoperable rail transport across member states in Europe. To this end, ERA maintains different registers with legal mandate covering different domains of railway (infrastructure, rolling stock, signalling, safety, humans, etc). Since 2020, ERA has adopted a data-centric organisation strategy, which includes leveraging semantic web technology to the different registers.
This talk will dive deeper into the current implementation as knowledge graph of two registers: the register of infrastructure (RINF) and the European Register of Authorised Types of Vehicles (ERATV). I will focus on the ontology development, the process management of the 50+ classes and 461 properties; as well as the SHACL rules (around 100). The talk will also showcase two applications consuming the knowledge graph – one for data retrieval for non SPARQL experts and the route compatibility check to answer if a certain railway vehicle can travel the route between two operational points. The talk will also highlight some challenges faced by a public authority when adopting a data-centric approach.
La présentation sera en français et commencera par un café offert par les Halles de l’IA de Université de Montpellier devant la salle 104 à 10h00.
LabéliséHallesIA
Montpellier IA CONNECT
17 février 2025 @ 14 h 00 min –
Salle du Conseil – Hôtel de Ville de Montpellier (1 place Georges Frêche)
Montpellier Métropole
Montpellier se positionne comme un acteur clé de l’innovation en France et renforce son engagement dans le développement de l’intelligence artificielle. Dans cette dynamique, nous avons le plaisir de vous inviter à l’évènement de lancement de l’association IA Montpellier Méditerranée !
Cet événement s’inscrit dans la continuité du sommet mondial sur l’IA et a pour ambition d’accélérer la transformation numérique et de structurer un écosystème fort et ambitieux.
Au programme :
• Lancement d’une grande enquête
• Présentation de projets innovants en IA
• Échanges sur les nouvelles pratiques
• Dévoilement du Comité Métropolitain du Numérique et de l’Intelligence Artificielle
Avec le soutien de la Ville et la Métropole de Montpellier, l’Université de Montpellier, du CHU de Montpellier, de la French Tech Méditerranée et de Digital 113, cet événement est l’occasion de valoriser les initiatives du territoire et d’inscrire Montpellier comme un moteur de l’innovation en intelligence artificielle.
LabéliséHallesIA
Rethinking Early Stopping: Refine, Then Calibrate
17 février 2025 @ 14 h 00 min –
Room 02.022, Building 5, St Priest campus
Machine Learning in Montpellier, Theory & Practice
Machine learning classifiers often produce probabilistic predictions that are critical for accurate and interpretable decision-making in various domains. The quality of these predictions is generally evaluated with proper losses like cross-entropy, which decompose into two components: calibration error assesses general under/overconfidence, while refinement error measures the ability to distinguish different classes. In this paper, we provide theoretical and empirical evidence that these two errors are not minimized simultaneously during training. Selecting the best training epoch based on validation loss thus leads to a compromise point that is suboptimal for both calibration error and, mostimportantly, refinement error. To address this, we introduce a new metric for early stopping and hyperparameter tuning that makes it possible to minimize refinement error during training. The calibration error is minimized after training, using standard techniques. Our method integrates seamlessly with any architecture and consistently improves performance across diverse classification tasks. [ https://isdm.umontpellier.fr/arxiv.org/abs/2501.19195 | https://isdm.umontpellier.fr/arxiv.org/abs/2501.19195 ]
Machine Learning in Montpellier, Theory & Practice
Rethinking Early Stopping: Refine, Then Calibrate
17 février 2025 @ 16 h 00 min –
Inria Montpellier, St-Priest Campus, Building 5, Room 02/022
Machine Learning in Montpellier, Theory & Practice
Machine learning classifiers often produce probabilistic predictions that are critical for accurate and interpretable decision-making in various domains. The quality of these predictions is generally evaluated with proper losses like cross-entropy, which decompose into two components: calibration error assesses general under/overconfidence, while refinement error measures the ability to distinguish different classes.
In this paper, we provide theoretical and empirical evidence that these two errors are not minimized simultaneously during training. Selecting the best training epoch based on validation loss thus leads to a compromise point that is suboptimal for both calibration error and, mostimportantly, refinement error. To address this, we introduce a new metric for early stopping and hyperparameter tuning that makes it possible to minimize refinement error during training. The calibration error is minimized after training, using standard techniques. Our method integrates seamlessly with any architecture and consistently improves performance across diverse classification tasks.
LabéliséHallesIA