2025 RHIC/AGS Annual Users' Meeting

Session

Workshop 6: AI/ML

21 May 2025, 09:00

Berkner Hall B (488)

Conveners

Workshop 6: AI/ML: Morning 1

• Yeonju Go (Brookhaven National Laboratory)

Workshop 6: AI/ML: Morning 2

• Yeonju Go (Brookhaven National Laboratory)

Description

Co-chair: Hannah Bossi, Tanner Mengel

412. Overview of AI/ML applications at RHIC

Jakub Kvapil (Los Alamos National Laboratory)

21/05/2025, 09:00

Artificial intelligence (AI) has a large transformative potential and is currently changing the industries around the globe. The deployment of cutting-edge AI techniques offers unparalleled opportunities to revolutionize data collection, reconstruction, and data analysis of the new era nuclear physics experiments. This talk will focus on the latest AI advancements and applications at the...

413. The Ecology of Uncertainty: Capturing Epistemic Doubt in Hadron Structure Analyses

Brandon Kriesten (University of Virginia)

21/05/2025, 09:30

Inverse problems are ubiquitous in hadron structure and tomography, where accurately characterizing uncertainties is crucial for unraveling new physics hiding within these uncertainties. In this new precision era of QCD, it is vital to create a translation between our physics and next generation AI/ML algorithms, using tools such as evidential deep learning and information-theoretic metrics to...

414. Preparing the Jet-AI/ML landscape for the EIC

Prof. Rithya Kunnawalkam Elayavalli (Vanderbilt University)

21/05/2025, 09:55

415. Machine learning on FPGAs for event selection

Hannah Bossi

21/05/2025, 10:45

416. Real-time Information Distillation on Novel AI Hardware

Yi Huang (Brookhaven national lab)

21/05/2025, 11:10

417. Generative AI for Jet Analysis and Detector Simulation

Shuhang Li (Columbia University)

21/05/2025, 11:35

We present denoising diffusion probabilistic models (DDPMs) as high-fidelity, AI-based generative surrogates for producing full-detector, whole-event simulations in heavy-ion experiments [1]. Trained on HIJING minimum-bias data propagated through the sPHENIX detector geometry with Geant4, DDPMs achieve roughly a hundredfold speedup over standard Geant4 simulations and exhibit superior fidelity...

418. Foundation Model for Tracking

Yihui Ren

21/05/2025, 12:00

The advent of large language models (LLMs) and foundation models has transformed the landscape of artificial intelligence. Unlike traditional AI methods that depend heavily on hand-crafted rules and heuristics, these models harness massive datasets and self-supervised learning to develop general-purpose representations. This paradigm enables them to adapt effectively to a wide range of...