IDEAS

Scientific Data Management and Visualization to Advance AI for Science

EVL’s work is part of a larger Argonne National Laboratory effort, PIONEER (Program for Intelligent Optimization for Next-generation Experiments, Explorations, and Research), formally Argonne Base: Scientific Data Management and Visualization to Advance AI for Science, funded under the DOE National Laboratory Program Announcement LAB 25-3520 and bringing together several collaborators across Argonne, partner universities and commercial partners (NVIDA, KitWare). The overall project supports the ongoing development and deployment of an AI-powered scientific assistant that transforms scientific exploration through intuitive, adaptive, and interoperable interaction with complex data. The work is organized in two thrusts, one in scientific data management and one in scientific data visualization. UIC participates in the thrust AI-Enhanced Visualization and Human-centered Data Interaction.

Research Hypothesis

Advanced AI-driven visualization and intelligent data exploration, when coupled with novel, efficient data representations, will dramatically accelerate scientific discovery by enabling real-time analysis of large-scale datasets and live experimental data.

Research Areas

To investigate this hypothesis, the project leverages exemplary data from a range of DOE domains to conduct research in three areas:

  • Human-centered exploration of scientific data
  • Efficient data representations for interactive exploration
  • AI-driven advanced rendering and visualization

UIC contributes to the design and development of an AI assistant for data exploration and a visualization workflow AI assistant.

Left: PIONEER all-hands meeting, February 2026, with the Visualization Workflow Assistant presented on the Arcade display wall at EVL and the Argonne team joining remotely. Right: an IDEAS board in SAGE3 gathering material on HACC cosmology and NekRS simulations for the AI assistants.

Team at UIC

  • Luc Renambot — UIC PI (EVL, Research Associate Professor, Computer Science)
  • Saeed Boorboor — Faculty collaborator and advisor (EVL, Assistant Professor, Computer Science)
  • Riccardo Bonfanti — Graduate student, advised by Saeed Boorboor

Project Details

Funding

This work is supported by the U.S. Department of Energy, Office of Science, via sub-award from Argonne National Laboratory.