3rd Workshop of the LIG SRCPR Axis (2022)
Systèmes Répartis, Calcul Parallèle et Réseaux
General information
- Organizers: Arnaud Legrand (POLARIS)
- When: May 12th, 9:00 – 12:30
- Who: everyone from the LIG is welcome
- Where: IMAG Amphitheater / BBB
- Questions for speakers: http://pads.univ-grenoble-alpes.fr/p/wax_srcpr_2022
9:00 - 9:20: Welcome
- Bring your own coffee
- 9:10: Introduction by Noel De Palma
Morning (9:20 - 12:30)
9h20 - 10:15: Session 1 (10-minute talks): Performance Evaluation
- Fabrice Rastello/Guillaume Iooss (CORSE): OpenOpt
- Sebastian Allmeier (POLARIS): Refined Mean Field Approximation for Interaction Model
- Martin Heusse (DRAKKAR): Capacity of LoRa
- Manuel Selva/François Broquedis/Florent Bouchez (CORSE): EasyTracker
- Yasmine Djebrouni (ERODS): Towards Bias Mitigation in Federated Learning
10:15 – 10:25: Coffee Break
10:25 – 11:10: Session 2 (10-minute talks): Climate and Societal Crisis
- Bernard Tourancheau (DRAKKAR): Quelle recherche en informatique dans le cadre des limites planétaires?
- Didier Donsez (ERODS): Présentation de l’equipex Terraforma : Prendre le pouls de la Terre
- Romain Couillet (POLARIS): La ``convivialité’’ comme métrique: les pistes d’une informatique low-tech
- Denis Trystram (DATAMOVE): Frugalité et apprentissage
11:10 – 11:20: Coffee Break
11:20 – 12:10: Session 3 (10-minute talks): Learning
- Lucas Nesi (POLARIS): Multi-Phase Task-Based HPC Applications: Quickly Learning how to Run Fast
- Danilo Santos Carastan (DATAMOVE): Towards convergence of AI and HPC: Large-scale AI simulation and multi-objective, hardware-aware AI development
- Kimang Khun (POLARIS): Reinforcement Learning for Markovian Bandits: Is Posterior Sampling more Scalable than Optimism?
- Simon Fernandez (DRAKKAR): Early Detection of Spam Domains with Passive DNS and SPF