SCOPE
Artificial Intelligence for the Electron Ion Collider
AI will be an essential part of future experiments like the Electron Ion Collider, a new $2B high-luminosity facility capable to collide high-energy electron beams with high-energy proton and ion beams that will be built at BNL in approximately 10 years from now to unlock the secrets of the "glue" that binds the building blocks of visible matter in the universe.
AI can provide new insights and discoveries from both experimental and computational data produced at user facilities.
This website includes AI/ML activities related to EIC that will characterize the different phases of its realization.
The AI4EIC paper summarizing the different activities and R&D projects, with an overview of the goals, approaches and strategies regarding AI/ML in the EIC community, can be found here.
AI4EIC: Background
Artificial Intelligence contributes to all phases of the Electron Ion Collider starting from the Design and R&D. The optimization of such large scale experiment is a complex problem characterized by multiple parameters and objectives like detector performance and costs. AI will provide insight on hidden correlations among the design parameters and will identify optimal tradeoff solutions in a multidimensional space of the objectives. The AI-supported Optimization of the Accelerator and Detector Design needs reliable Simulations followed by Reconstruction and Analysis, all areas covered in the AI4EIC workshop. Another important activity in EIC is Streaming Readout, aiming at a continuous readout of all detector signals without requiring a "trigger", with data selection realized in software, furthering the convergence of online and offline analyses and allowing for faster data quality monitoring, calibration and alignments. EIC will be one of the first automated experiments where AI will be largely applied for Accelerator and Detector control. EIC will be operating in 2030's, and by then AI may also leverage on technologies that are currently Computing Frontiers.
List of areas (in progress)
- Accelerator and Detector Design
- Simulations
- Analysis and Reconstruction
- Accelerator and Detector Control
- Streaming Readout
- Computing Frontiers
- Theory and Phenomenology
The EIC detector can be one for the first large-scale detector to be designed with the assistance of AI. Within the AI4EIC WG we are developing tools to facilitate the interactive "navigation" of the Pareto solutions for the EIC design
https://ai4eicdetopt.pythonanywhere.com
https://cfteach.github.io/nnpss
AI4EIC: Workshops
The AI4EIC is a series of workshops.
The 1st workshop, was dedicated to experimental applications of A.I. for the Electron Ion Collider and took place on September 7-10 2021 at CFNS during the design and R&D phases.
The second AI4EIC workshop has been hosted by William & Mary on October 2022.
The third AI4EIC workshop is hosted at Catholic University of America, Washington D.C., during November 28-December 1, 2023.
A full list of past and future events can be found at https://eic.ai/events
AI4EIC: Outreach
AI in our society will be the economic driver of the next decade when EIC will be operating. Educational activities are aimed at disseminating AI in the EIC community. Hackathons can be built around specific challenges for EIC and help identifying strategies, architectures and algorithms that will benefit the EIC physics program.
Collaborate with us
Click here for more information
mailing list: eicug-software-ai@eicug.org
join the ai4eic slack channel
Copyright © 2021 Artificial Intelligence for the Electron Ion Collider - All Rights Reserved.
ai4eic
This website uses cookies.
We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.