Machine Learning Applications for Particle Accelerators
Date:
Tuesday, February 27 through Friday, March 2, 2018
Location:
SLAC National Accelerator Laboratory
Overview:
We are pleased to announce the first ICFA mini-workshop on Machine Learning, to be held at SLAC National Accelerator Laboratory in Menlo Park, California, from February 27 to March 2. The goal of this workshop is to help build a world-wide community of researchers interested in applying machine learning techniques to particle accelerators. The workshop will be split into four sequential topics:
- Tuning/optimization/control
- Prognostics/alarm handling/anomaly-breakout detection
- Data analysis
- Simulations/modeling
Talks will include both accelerator physicists and computer scientists. This workshop has the following goals:
- Collect and unify the community’s understanding of the relevant state-of-the-art ML techniques.
- Provide a simple tutorial of machine learning for accelerator physicists and engineers.
- Seed collaborations between laboratories, academia, and industry.
- Author a whitepaper explaining the current opportunities for ML techniques in particle accelerators, with a few illustrative examples. This whitepaper should explain why now is the time for the community to fully embrace ML alongside optimization as the modern way to aid particle accelerator design and operation.
Given the early state of machine learning at accelerator facilities, we will place a heavy emphasis on discussions, collaboration planning, and poster sessions, with only a few general talks. Those interested are encouraged to join a day of tutorials on Tuesday, February 27. The main workshop will begin on Wednesday, February 28. Due to limited space, this workshop is by invitation only. Please contact the organizers if you are interested to attend.
Registration is now closed.
Start time | Title | presenter |
| 7:30-9:30 AM | Breakfast | |
| 9:00 AM -12:00 PM | ML Tutorial: Morning Conveners: Auralee Edelen (CSU) Dr Christopher Mayes (SLAC National Accelerator Laboratory) Daniel Bowring (Fermilab) Gianluca Valentino | 9:00 AM | Logistic Regression, Neural Networks 10:30 AM | Cofee 11:00 AM | Neural networks
|
| 12:00 PM -1:00 PM | Provided Lunnch Buffet | |
| 1:00 PM - 2:30 PM | ML Tutorial:Afternoon | 1:00 PM | Unsupervised Learning (Clustering), Toy Optics model 2:30 PM | Coffee |
| 3:00 PM-5:00 PM | Ocelot Tutorial | Convener: Sergey Tomin |
Start time | Title | presenter |
| 7:30-9:30 AM | Provided Breakfast | |
| 9:00-9:20 AM | Welcome | Speaker: Prof. Tor Raubenheimer (SLAC) |
9:20 AM -12:00 PM
| Facility needs Convener: Kevin Li
| 9:20 AM | Facility needs: Hadron Synchrotrons: Operational challenges at the LHC 9:40 AM |Facility needs: XFELs Speaker: Raimund Kammering (DESY) 10:00 AM | Facility needs: Synchrotrons Speaker: Xiaobiao Huang 10:20 AM | Coffee 10:50 AM | Fault detection and alarm systems for the CERN Technical Infrastructure Speaker: Jesper Nielsen (CERN) 11:10 AM | Facility needs - or chances - seen from the other side Speaker: Arno Candel 11:30 AM | Discussion Speaker: Kevin Li |
| 12:00 PM-1:00 PM | Provided Lunch Buffet | |
1:00 PM - 6:00 PM
| Tuning Convener: Auralee Edelen
| 1:00 PM | Experience with FEL taper tuning using reinforcement learning and clustering 1:20 PM |Introduction to Bayesian Optimization Speaker: Johannes Kirschner 1:40 PM | Experience at SLAC with Bayesian Optimization using Gaussian Processes Speaker: Jopseph Duris 2:00 PM | Poster Blitz 2:20 PM | Coffee 3:20 PM | Experience with tuning at XFEL Speaker: Sergey Tomin Slides (PDF) | (PPT) 3:40 PM | Experience with tuning at FERMI@Elettra Speaker: Giulio Gaio (FERMI@Elettra) 4:00 PM | General experience with online optimization Speakers: Alexander Scheinker, Scheinker Alex 4:20 PM | Sloppy and Genetic Algorithms for Low-Emittance Tuning at CESR Speaker: Ivan Bazarov 4:00 PM | Discussion: Tuning Speaker: Auralee Edelen (CSU) |
| 6:00 PM-9:00 PM | Reception at Dutch Goose | |
Start time | Title | presenter |
| 7:30-9:00 AM | Breakfast | |
| 9:00-12:00 PM | Simulations and Modeling Convener: llya Agapov(DESY) | 9:00 AM | Experience with Model Predictive Control and Model-based Reinforcement Learning using Neural Networks Speaker: Auralee Edelen 9:20 AM | Forecasting of Beam Interlocks in High Intensity (Hadron) Accelerators Speaker: Andreas Adelmann 9:40 AM | Light Source Simulations Speaker: Ilya Agapov (DESY) Slides (PDF) | (PPT) 10:00-11:00 AM | Poster Session and Coffee 11:00 AM | Neural Network Modeling and Virtual Diagnostics at FAST Speaker: Jonathan Edelen 11:20 AM | GANs for Simulation in HEP Speaker: Luke de Oliveira 11:40 AM | Discussion Speaker: Ilya Agapov (DESY) |
| 12:00 PM-1:00 PM | Lunch | |
| 1:00 PM- 3:00 PM | Prognosics Convener: Rasmus Ischebeck | 1:00 PM | Machine learning for anomaly detection in large distributed accelerator systems Speaker: Pieter van Trappen 1:20 PM | Beam loss plane recognition for the LHC Speaker: Gianluca Valentino Slides (PDF) | (PPT) 1:40 PM| Detection of bad BPMs Speaker: Elena Fol Slides (PDF) | (PPT) 2:00 PM | Discussion 2:30 PM |Coffee |
| 3:00 PM - 5:00 PM | Tour | |
| 5:30-8:00 PM | Meyer-Buck House Reception | |
Reception at the Provost's house. Parking is at the Hewlett Foundation next door.
| Start time | Title | presenter |
| 7:30-9:30 AM | Breakfast | |
| 9:00-11:00 AM | Data Analysis Convener:Daniel Bowring(Fermilab) | 9:00 AM | Machine learning application on the investigation of the micro-bunching instability at storage rings Speaker: Mr. Tobias Boltz(KIT) 9:20 AM | Non-Parametric Density Estimators for the Measurements of Ionization Cooling Speaker: Tanaz Angelina Mohayai (Illinois Institute of Technology) 9:40 AM | Data mining at HIPA Speaker: Mr. Jochem Snuverink (Paul Scherrer Institute) |
| 10:00-11:00 AM | Poster Session and Coffee | |
| 11:00 AM-12:00 PM | Summary | Auralee Edelen(CSU) Daniel Bowring (Fermilab) Ilya Agapov (DESY) Kevin Li (CERN) Dr Rasmus Ischebeck (PSI) |
| 12:00 PM - 1:30 PM | Lunch | |
| 1:30-3:00 PM | White paper and collaboration planning | Speaker: Dr Christopher Mayes (SLAC national Accelerator Laboratory) |
