US ATLAS Machine Learning Training 2023

→ US/Pacific
59/3101 (Building 59) (LBNL)

59/3101 (Building 59)

LBNL

1 Cyclotron Rd, Berkeley, CA 94720
Aishik Ghosh (University of California Irvine (US)), Elham E Khoda (University of Washington (US)), Ben Nachman (Lawrence Berkeley National Lab. (US))
Description

 

The second iteration of the US ATLAS Machine Learning (ML) training event will be hosted at Lawrence Berkeley National Laboratory. All talks and tutorials will be given in person. The workshop is open to all ATLAS collaborators.

Overview:

We will introduce fundamental concepts of machine learning accompanied by hands-on tutorials of the essential open-source ML packages. The program will cover particle physics specific use cases and deployment of the trained models in Athena/FPGAs, with lots of hands-on examples. There will be invited talks from ATLAS members who have previously deployed ML for different tasks in ATLAS, as well as experts from CMS adn beyond. Use cases in other experiments and other scientific domains will also be discussed to provide a glimpse into the larger ML4Science world.

Attendees can expect to gain an overview of the broad range of current and potential ML applications in ATLAS, and also learn some of the particle physics specific tricks that an ML practitioner picks up from experience. We will try to address the typical ML questions that often come up in ATLAS meetings.

Tentative topics to be covered:

  • Introduction to Machine Learning
  • Introduction to standard open-source ML packages like Scikit-learn, XGBoost, Keras and Pytorch (hands-on)
  • Overview of ML in particle physics
  • Practitioners guide to handling particle physics datasets
  • Generative Models (hands-on)
  • Uncertainty treatment
  • Unfolding
  • Simulator Based Inference
  • Exploiting symmetries in physics data (hands-on)
  • Graph Neural Networks in particle physics (hands-on)
  • Transformers and LLMs
  • Differentiable Programming (hands-on)
  • Deploying NNs in C++ and python: ONNX Runtime (hands-on)
  • Deploying NNs on FPGA: Quantization, Model prunnning and compression, HLS4ML (hands-on)

Please look at the time table to for the full agenda.

Industry Talk:

This year we will have Chen Luo from Amazon talk to us about his experience working in industry.

Computing Resources:

In-person participants will be guaranteed computing resources for the hands-on sessions thanks to NERSC. Virtual participants will be giving access on a first come first serve basis. 

Networking:

The inaugural program led to new collaborative proejcts in ATLAS. This training program will again be a platform for young ML enthusiasts to connect with one another and with veteran ML experts in ATLAS.

Tutorial git:

Github link: https://github.com/usatlas-ml-training/lbnl-2023/tree/main 

Zoom: 

Discussions:
Join the slack workspace to discuss and ask questions about the tutorials, particularly for remote participants.

slack joining link:

Zoom
Contact
Participants
  • Andrej Saibel
  • Bradley Duane Hutchinson
  • Camila Pazos
  • Christian Weber
  • CONNOR WAITS
  • Danijela Bogavac
  • Diptaparna Biswas
  • Egor Antipov
  • Elham E Khoda
  • Hui-Chi Lin
  • Ilkay Turk Cakir
  • Jack Harrison
  • Jason Gombas
  • Joni Pham
  • Juan Carlos Cardenas Jr
  • Judita Mamuzic
  • Juerg Beringer
  • Luis Felipe Gutierrez
  • Luke Martin Vaughan
  • Marco Lisboa Leite
  • Mario Campanelli
  • Mason Ray Housenga
  • Matthew Drnevich
  • Meng-Ju Tsai
  • Mohamed Belfkir
  • Mohamed Reda MEKOUAR
  • Ning Ni
  • Qiuping Shen
  • Rebecca Hicks
  • Ricardo Barrué
  • Rodrigo Estevam De Paula
  • Saad Mohiuddin
  • Stergios Kazakos
  • Tom Cheng
  • Wasikul Islam
  • Zackary Lee Alegria
  • Zekeriya Uysal
  • Zubair Bhatti
  • +54