AI tutorials

From GlueXWiki
Revision as of 15:46, 11 February 2025 by Ijaegle (Talk | contribs)

Jump to: navigation, search

Purpose

The overall topic is: „Best practice for AI in nuclear physics applications“. We wish to cover the following items via plenary talks and/or interactive tutorials:

  • Feature engineering --> Feature normalization, correlation coefficients, feature selection, etc.
  • Overfitting --> Dropout layers, weight regularization, etc.
  • Performance evaluation --> ROC-Curve, confusion matrix, accuracy, loss curves,...
  • Visualization --> How to properly present the performance of a model ? What are "good" diagnostic plots ?
  • Model deployment --> How to use a model within the GlueX analysis software
  • Data fed into models --> What data sets are used ? Numerical, vs. Images, raw data vs. clean data,...
  • Tools made available by the data science department
  • Optional, depending on time: HPO --> Tune the parameters of your model
  • Optional, depending on time: Bookkeeping of models via MLFlow

Location and Time

The workshop will take place at:

DATES: February 18 (all day) - 19 (morning only), 2025

LOCATION: CEBAF Center F113

 

Remote Participation


 

References

 

Workshop Software

PID example Notebook

Agenda

AI Tutorials

Feb 18

  • 09:00 Welcome (5min)
  • 09:05 Brief introduction to Machine and Deep Learning (1h35)
  • 10:40 Break (20 min)
  • 11:00 Notebook setup (15min)
  • 11:15 Data preparation and feature engineering (1h15)
  • 12:30 Lunch (1 h)
  • 13:30 Setting up a model and training it (1h15)
  • 14:45 Final performance evaluation (1h15)
  • 16:00 Break (20min)
  • 16:30 HYDRA (45min)

Feb 19

  • 09:40 TMVA (30min)
  • 10:10 TMVA (30min)
  • 10:40 Break (20min)
  • 11:00 Graph Neural Network (45min)

 

Registration

Please add your name to the list of attendees below. No formal registration or registration fee is required.

{ | border="1" cellpadding="2" valign="top" style="text-align: center;" !width="180"| Name !width="200"| Home Institution !width="200"| Level !width="200"| Participate at JLab |- | ig || JLab || Staff || Yes |- | Daniel || JLab || Staff || Yes |- | Gyang || Virginia Tech || Student || Yes |- | Karthik || William and Mary || Postdoc || No |- | Zachary Baldwin || Carnegie Mellon University || Graduate Student || Yes |- | Nizar Septian || Florida State University || Student || Yes |- | Alex Austregesilo || JLab || Staff || Yes |- | Drew Smith || JLab || Postdoc || Yes |- | Sean Dobbs || FSU || Faculty || Yes |- | Will Imoehl || Carnegie Mellon University || Postdoc || No |- | Farah Afzal || Ruhr University Bochum || Faculty || Yes |- | Peter Hurck || Glasgow || Faculty || Yes |- | Lawrence Ng || JLab || Postdoc || Yes |- | Albert Fabrizi || University of Mass. Amherst || Graduate Student || Yes |- | Shannen Graham-Howard || University of Mass. Amherst || Graduate Student || Yes |- | Jiawei Guo || Carnegie Mellon University || Graduate Student || Yes |- | Boris Grube || JLab || Staff || Yes |- |Churamani Paudel || FIU || Postdoc || No |- |Vitor Shen || Ruhr University Bochum || Graduate Student || No |- }  

Recording