This US ATLAS Machine Learning (ML) training event will be hosted at Lawrence Berkeley National Laboratory (hybrid mode). All talks and tutorials will be given in person, and in person attendance is encouraged for participants, if they are able. 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. Few 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 (hands-on)
- Introduction to Neural Networks with Keras (hands-on)
- Overview of ML in particle physics
- Practitioners guide to handling particle physics datasets
- Uncertainty treatment
- Unfolding
- Simulator Based Inference
- Exploiting symmetries in physics data (hands-on)
- Graph Neural Networks in particle physics (hands-on)
- Anomaly detection (hands-on)
- Deploying NNs in C++: ONNX Runtime (hands-on)
- Deploying NNs on fast hardware : HLS4ML (hands-on)
Please look at the time table to for the full agenda.
Industry Talk:
For the industry talk, Dr. Jaideep Pathak will discuss his journey transitioning from academia to research work at NVIDIA for weather predictions with ML.
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:
This training program also aims to 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
Zoom links:
https://cern.zoom.us/j/67597522045?pwd=Ty83d1VRWFF2OGVJVm9PaXA5bEtWQT09
Discussions:
Join the slack workspace to discuss and ask questions about the tutorials, particularly for remote participants.
slack joining link: https://join.slack.com/t/atlasmltraining/shared_invite/zt-1daz1y3z4-wihELkZx2MdNb9CfyjzoPA
