IRIS-HEP Fellow: Oleksii Kiva
Fellowship dates: Jul – Oct, 2022
Home Institution: Igor Sikorsky Kyiv Polytechnic Institute
Project: Developing an automatic pruning utility for statistical models in HistFactory format
In pyhf, large-scale statistical models employed in HEP experiments are constructed using a modular approach to build a parametrized family of complex probability density functions from more primitive conceptual building blocks. It is often useful to make the model more lightweight in order to speed-up the derivation of maximum-likelihood estimates of its parameters. What's already available in pyhf is only helpful if one manually decides and specifies exactly what blocks to remove ('prune') from the statistical model. The goal of this project is to devise, implement, document and integrate into the pyhf library framework a tool that will automatically decide how to reduce the statistical model in HistFactory format, given its pyhf-specification and some ‘pruning’ criteria for the blocks.More information: My project proposal
Mentors:
-
Alexander Held (University of Wisconsin-Madison)
Current Status
Bachelor of applied mathematics
Contact me: