AI/ML for Nuclear Physics HUGS 2025

The HUGS 2025 - Artificial Intelligence for Nuclear Physics

(May 27 - Jue 13, 2025)

This is the landing page of the AI lectures of HUGS 2025

Synopsis: The HUGS2025 lectures on Artificial Intelligence for Nuclear Physics are a focused 6-hour program designed to introduce physics graduate students to the fundamentals of artificial intelligence and machine learning (AI/ML), and their applications in nuclear physics (NP); many examples are taken from the Electron Ion Collider (EIC) project and Jefferson Lab (JLab) experiments.

The course aims to equip students with a basic understanding of AI/ML basics, and how these techniques can be utilized to interpret and analyze NP data.

The students will gain a practical understanding of AI/ML concepts through hands-on exercises using simulated NP data. The lectures serve as a launchpad for those aspiring to integrate AI/ML techniques into their NP research while also offering valuable insights into some of the latest AI/ML initiatives ongoing at both the JLab and EIC.

This course is based on the following references [ABLN22, BDS+22, Fan21, FN21, K+21, MBW+19]

📚 2025 Lectures

⚠️ Note: Course material will appear below.

Lectures

• (Lec 1, part A) A Brief Introduction to AI/ML
• (Lec 1, part B) AI/ML in Nuclear Physics
• (Lec 2) CNN for Physics by K. Suresh
• (Lec 3) Multi-Objective Optimization by K. Suresh

Tutorials

• (Lec 1) Deeply Learning Deep Inelastic Scattering
• (Lec 2) CNN - Part 1
• (Lec 2) CNN - Part 2
• (Lec 2) Dataset
• (Lec 3) Design with Multi-Objective Bayesian Optimization
• (Lec 3) Design with Multi-Objective Evolutionary Optimization

(Old) Lectures

• (Lec 1) AI/ML for Nuclear Physics - A High-Bias, Low-Variance Introduction
• (Lec 2) Leveraging AI/ML for the future Electron Ion Collider
• (Lec 2, extra) AI/ML activities at Jefferson Lab - Paving the way for the Electron Ion Collider
• (Lec 3) ePIC at EIC – The First Large-Scale Experiment Designed with the AI Assistance

(Old) Tutorials

• (Tut 1) "Pre-flight"
• (Tut 2 - part a) AI-assisted design with MOEA
• (Tut 2 - part b) AI-assisted design with MOBO
• AI4EIC tutorials

[Pre-flight]

• Intro to Modeling, Linear Regression (lecture slides)
• 1. Brief course on git
  • 1.1. Version Control System (VCS)
  • 1.2. What is Git?
  • 1.3. Configuring your git
• 2. Exercises on git
  • 2.1. Exercise 1: Basic Git Commands
  • 2.2. Exercise 2: Commits, push and branches
  • 2.3. Advanced exercises
  • 2.4. GitHub Actions and Workflow Tutorial
• 3. Installing Visual Studio Code
• 4. Installing Extensions
• 5. Signing up and Installing GitHub Copilot
• 6. Introduction to High Performance Computing
  • 6.1. High Performance Computing
  • 6.2. SSH into a Front-End Node Using Proxy Jump
  • 6.3. William & Mary Research Computing: File Systems Tutorial
  • 6.4. Introduction to SLURM

Other References

• Introduction to Data Science (beginners)
• Interactively Navigate the Pareto Front of the EIC Tracker
• EIC Overview

Datasets / Modules

• (Lec 1) all-h1-rapgap.root
• (Lec 1) intro_Data.py
• (Lec 2) Dataset

Additional resources

• References

📚 Past Editions

• 2023 HUGS Lectures on AI/ML for Nuclear Physics

Credits: Material on git, VS-Code, and HPC from AID2E