Testing new physics models with global comparisons to collider measurements: the Contur toolkit

Abstract

Measurements at particle collider experiments, even if primarily aimed at understanding Standard Model processes, can have a high degree of model independence, and implicitly contain information about potential contributions from physics beyond the Standard Model. The CONTUR package allows users to benefit from the hundreds of measurements preserved in the RIVET library to test new models against the bank of LHC measurements to date. This method has proven to be very effective in several recent publications from the CONTUR team, but ultimately, for this approach to be successful, the authors believe that the CONTUR tool needs to be accessible to the wider high energy physics community. As such, this manual accompanies the first user-facing version: CONTUR v2. It describes the design choices that have been made, as well as detailing pitfalls and common issues to avoid. The authors hope that with the help of this documentation, external groups will be able to run their own CONTUR studies, for example when proposing a new model, or pitching a new search.

Picking the low-hanging fruit: testing new physics at scale with active learning

“…The current CONTUR methodology [1,3] obtains exclusion results for all the desired points in a new-physics model's parameter space by generating collider events using the Herwig event generator [5] with a Universal FeynRules Object [6], used to describe the BSM model. Generated signal events are passed through the analysis logic of the preserved RIVET routines [7], and the results distribution are compared to the observed results for LHC analyses.…”

“…There are several metrics which can be used to quantify this, as discussed in Section 2.4. Other performance metrics are also calculated for the testing and training pools, and determine the stopping conditions which dictate whether the ORACLE should continue sampling or provide a final prediction 1 . A summary of this procedure is provided in Figure 2.…”

“…Hundreds of measurements have already been made in a wide variety of final states. New tools [1] can harness the power of these measurements to efficiently discard those theories that are incompatible with past observations. Even though historic measurements were primarily aimed at understanding Standard Model processes, they implicitly contain information about potential contributions from physics beyond the Standard Model.…”

“…To achieve this goal, the CONTUR project [1,3] has formalised a methodology for testing a wide array of candidate new-physics models. It simulates model behaviour by generating events using Monte Carlo methods, and compares these results with hundreds of differential cross-section measurements from ATLAS, CMS and LHCb.…”

BSM Developments and Tools

“…A non-exhaustive list of these include MadAnalysis 5 [7][8][9][10][11][12], which can be used with a detector simulation or transfer functions and contains around 40 run I and II analyses; CheckMATE [13,14], which contains over 50 run I and II analyses, and was recently extended to support long-lived particle searches in addition to prompt particle searches; ColliderBit [15], which includes its own detector modelling in the form fast 4-vector smearing (Buckfast), and has a database of around 40 analyses, from runs I and II ; and Rivet [17,18], which has currently "only" 30 BSM analyses, but boasts a huge library of SM analyses, upwards of 800. In fact, it has been shown that SM inclusive measurements also set strong constraints on new physics, and the tool Contur has been developed for this purpose [19][20][21][22][23]. A comparison of the performance of these tools for a specific CMS search for supersymmetric (SUSY) particles [24] was performed [25] and strong agreement was seen across all included tools.…”