Workshops Archive

Scale to See: Scalable AI for Scientific Imaging and Spatiotemporal Data


Workshop: Scale to See: Scalable AI for Scientific Imaging and Spatiotemporal Data

Authors: Xiao Wang (Oak Ridge National Laboratory (ORNL)), Mohamed Wahib (RIKEN Center for Computational Science (R-CCS)), Charles Bouman (Purdue University), and Yuankai Huo (Vanderbilt University)

Abstract: Scale to See: Scalable AI for Scientific Imaging and Spatiotemporal Data is a half-day workshop at SC that brings together researchers at the intersection of high performance computing (HPC), artificial intelligence (AI), large-scale scientific imaging and spatiotemporal data. Scientific imaging and spatiotemporal data—from climate and earth systems to medical imaging, materials science, and cosmology—are increasingly high-dimensional, multimodal, and time-evolving, demanding AI methods that are not only accurate but also scalable, physics-aware, and deployable on leadership-class HPC systems. This workshop focuses on AI methods and systems explicitly co-designed for extreme scale, including HPC-aware model architectures, scalable training and inference pipelines, physics-integrated learning algorithms, and efficient computing techniques for high-dimensional data. Through peer-reviewed papers, invited talks, and panel discussions, the workshop aims to set a research agenda, foster interdisciplinary collaboration, and build a community around AI that operates natively at HPC scale to transform scientific imaging and spatiotemporal analysis.

Website: https://sites.google.com/view/scale-to-see-workshop/home


Back to Scale to See: Scalable AI for Scientific Imaging and Spatiotemporal Data Archive Listing

Back to Full Workshop Archive Listing