OmniFold: A Method to Simultaneously Unfold All Observables

TL;DR

OmniFold is a novel machine learning-based unfolding method that iteratively reweights simulated data to correct for detector effects, enabling simultaneous, high-dimensional analysis of all observables in collider data.

Contribution

It introduces a unbinned, high-dimensional unfolding technique that leverages all available information and allows for simultaneous measurement of multiple observables.

Findings

Successfully applied to jet substructure data from LHC

Outperforms traditional binned unfolding methods

Enables measurement of all observables simultaneously

Abstract

Collider data must be corrected for detector effects ("unfolded") to be compared with many theoretical calculations and measurements from other experiments. Unfolding is traditionally done for individual, binned observables without including all information relevant for characterizing the detector response. We introduce OmniFold, an unfolding method that iteratively reweights a simulated dataset, using machine learning to capitalize on all available information. Our approach is unbinned, works for arbitrarily high-dimensional data, and naturally incorporates information from the full phase space. We illustrate this technique on a realistic jet substructure example from the Large Hadron Collider and compare it to standard binned unfolding methods. This new paradigm enables the simultaneous measurement of all observables, including those not yet invented at the time of the analysis.