INNFER (Invertible Neural Networks for Extracting Results) is a framework for performing simulation-based inference in a frequentist statistical setting. It enables multidimensional, unbinned likelihood fits by leveraging machine learning techniques to model probability densities.

INNFER uses artificial intelligence to learn the probability density function:

\[p(\vec{x} \mid \vec{\theta}, \vec{\nu})\]

where:

  • $\vec{x}$ = reconstructed variables
  • $\vec{\theta}$ = parameters of interests
  • $\vec{\nu}$ = nuisance parameters

This approach provides an optimal and statistically rigorous way to perform high-dimensional data analyses while maintaining the interpretability of classical frequentist methods. A full description of the statistical methods used is discussed here.

The framework is developed at Imperial College London for initial use performing statistical analysis on high-energy physics data collected by the CMS experiment at the Large Hadron Collider (LHC). If you have any questions about the repository contact George Uttley at george.peter.uttley@cern.ch.

Contributors: George Uttley, Nicholas Wardle, Ye He


Next: Setup Guide.