- Dates: 19-21 June 2024
- Venue: Occidental Atenea Mar, Barcelona, Spain
The program comprises thematic plenary sessions and parallel workshop sessions organized around multiple themes, each featuring lightning talks, panels, and group discussions. These sessions aim to facilitate the formation and deepening of collaborations, explore opportunities to coordinate efforts, and strengthen collaborations within the European AI, HPC, and Science communities. The objective of the TPC workshop program is to advance the building, evaluating, and use of large-scale AI models for science and engineering applications.
19-June-2024 Wednesday
09:00-14:00 Tutorial: Large Language Models (LLMs) for Science and Engineering Applications (optional)
14:00-15:00 Welcome Lunch
15:00-17:45 Opening Plenary
- Welcome from the Barcelona Supercomputing Center
- Rick Stevens (Associate Laboratory Director, Argonne National Laboratory): A Vision for AI in Science
- Satoshi Matsuoka (Director, RIKEN Center for Computational Science): AI for Science at Riken and Japan – Fugaku-LLM, Riken TRIP-AGIS, and FugakuNEXT
- Mateo Valero (Director, Barcelona Supercomputing Center): AI and Architectures at BSC
- Introduction to the Trillion Parameter Consortium and How to Optimize Your Participation this Week! (Charlie Catlett, Argonne National Laboratory)
19:00 Welcome Reception
20-June-2024 Thursday
09:00-10:30 Plenary: Major International and National AI Initiatives and Vision
- Jérôme Bobin (CEA) – Update on the International Post-Exascale initiative (InPex)
- International Panel on AI Funding Initiatives
- Moderator: Fabrizio Gagliardi (BSC) with panelists highlighting EU, US, and Japanese initiatives.
- Speakers include: Daniela Petkova (EU), Satoshi Matsuoka (Japan), Rick Stevens (US)
10:30-11:00 Coffee Break
11:00-13:00 Plenary: Insights from Building Large-Scale AI Models for Science
- Ian Foster (Argonne / University of Chicago) – AuroraGPT: Building an Interdisciplinary Foundation Model
- Sampo Pyysalo (University of Turku) – Large models for Small Languages
- Jenia Jitsev (JSC/LAION) – Open Foundation Models: Reproducible Science of Transferable Learning
- Marko Grobelnik (Jozef Stefan Institute) – Open GenAI Challenges with Large-Scale Models
13:00-14:00 Lunch
14:00-15:30 Parallel Breakout Sessions
- Model Architecture and Performance Evaluation (1 of 2)
- Bioinformatics/Cancer/Drugs (1 of 2)
- Growing and Training the Community
15:30-16:00 Coffee Break
16:00-17:30 Parallel Breakout Sessions
- Data, Training Workflows, LLM Strategies (1 of 2)
- Bioinformatics/Cancer/Drugs (2 of 2)
- Early Experiences in Using LLMs for Scientific Software Use Cases
- AI Hardware Acceleration Strategies at Scale
21-June-2024 Friday
09:00-10:30 Parallel Breakout Sessions
- Data, Training Workflows, LLM Strategies (2 of 2)
- Skills, Safety, and Trust Evaluation
- Model Architecture and Performance Evaluation (2 of 2)
- LLMs for Healthcare
10:30-11:00 Coffee Break
11:00-13:30 Closing Plenary: Scale, Trust, and Responsible AI
- Franck Cappello (Argonne) – AuroraGPT/Evaluation of AI Assistant for Science: Critical and non-Trivial
- Mikel Rodriguez (Google DeepMind) – Emerging challenges in Securing Frontier AI Systems
- Uljan Sharka (iGENIUS) – Composite AI: Cutting Through the Hype
- Rio Yokota (TiTech) – LLM Pre-training Efforts in Japan
- TPC Wrap-Up and Path Forward (Charlie Catlett, Argonne National Laboratory and University of Chicago)
- Closing remarks from BSC Hosts (Mateo Valero, Fabrizio Gagliardi)
13:45 Workshop Concludes