Program

SciFM24 at the University of Michigan

Breakfast

8:00 AM

8:30 AM

Charting the course for SciFMs: Themes & Questions, K. Duraisamy (UM)

Emerging Paradigms in SciFM

9:00 AM

9:00 AM: Foundational Methods for Foundation Models for Scientific Machine Learning, Michael W. Mahoney (UC Berkeley, ICSI, LBNL)

9:40 AM: Scaling Laws of Formal Reasoning, Sean Welleck (CMU)

10:10 AMNext-generation Frameworks for AI and Science, and Trust, Payel Das (IBM)

10:40 AM Coffee Break

Panel Discussion: Big Questions for SciFM

11:00 AM

12:00 PM

Lunch, and Contributed posters

Frontiers of SciFM:

1:30 PM

1:30 PM: Trillion Parameter Foundation Models as Discovery Accelerators, Ian T. Foster (ANL, UChicago)

2:05 PM: Neural Operators: AI Accelerating Scientific Understanding, Animashree Anandkumar (Caltech)

2:40 PM: Fundamental Physics and Foundational Models for the Forecasting and Optimization of Complex Systems, Petros Koumoutsakos (Harvard)

10:40 AM Coffee Break

3:30 PM: University Laboratory Partnerships in the AI Era, Jason Pruet (LANL)

4:00 PM

Panel Discussion: Funding and Venture Ecosystem

Award Ceremony and Closing Remarks

5:00 PM

5:30 PM

Adjourn

8:30 AM

Opening Remarks & Conference Goals: Karthik Duraisamy

Applications of LLMs in Science:

09:00 am : Joint Language and Molecule Representation for Drug Discovery, Heng Ji (UIUC)

09:25 am : Foundation Models for Enabling Rational Design of Biological Systems, Arvind Ramanathan (ANL)

09:50 am : Scaling up Materials Foundation Models, Venkat Viswanathan (UM)

9:00 AM

10:15 AM Break

10:30 AM

Hands-on introduction to Scientific LLMs [TPC]

Tutorial material: https://github.com/ramanathanlab/SciFM24-Tutorial/tree/main

Lunch [Michigan League: Michigan and Kalamazoo Rooms] (download map)

11:30 AM

1:00 PM

Building SciFMs & SciLLMs:

The central epistemological debate in scientific ML: what can be learned from data, what must be encoded as constraint, and what lies beyond the reach of current architectures? Speakers bring perspectives from turbulence modeling, climate emulation, materials design, and HPC-scale simulation.

Track 1: LLM Tools for Science [TPC] at Rackham Amphitheater

• Retrieval augmented generation

• Agent-based systems

• Building toolchains and surrogates

• Genome Scale language models

• Scaling experiments on Supercompute

Track 2: LLM Tools for Mathematics [Sean Welleck] at East Conference Room

• Open Language for Mathematics

• Neural Theorem Proving

Break

3:00 PM

Session 1: Office hours for Advanced users [TPC]

3:15 - 5:00PM

Session 2: Deploying GenAI models with Nvidia Inference Microservices [NVIDIA]