Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Searching for HWW Anomalous Couplings with Simulation-Based Inference

Marta Silva, Ricardo Barrué, Inês Ochoa, Patricia Conde Muíño

Abstract

Understanding the source of the universe's asymmetry between matter and antimatter is one of the major open questions in particle physics. In this work, the sensitivity of novel machine-learning-based inference techniques to CP-odd and CP-even $HWW$ anomalous couplings is studied in the $\\WH \rightarrow \ell νb\bar{b}$ channel ($\ell = e, μ$), within the Standard Model Effective Field Theory (SMEFT) framework. Two machine-learning simulation-based inference (SBI) methods are explored: a per-event likelihood-ratio estimator, which directly approximates the ratio of probability densities between competing hypotheses, is benchmarked against a per-event optimal-observable estimator optimized for sensitivity to the parameters of interest. Both approaches are also compared to traditional summary statistics, in this case histograms of kinematic and angular observables, as commonly used in experimental analyses. SBI methods provide tighter constraints than one-dimensional summary statistics, though their performance is comparable to two-dimensional histogram analysis. The optimal-observable approach remains promising for its ability to probe multiple couplings simultaneously. Restricting the analysis to a region of high $S/B$ also enhances sensitivity to CP-odd operators while preserving sensitivity to CP-even operators, which histogram analyses often lose. Although the likelihood-ratio estimator sometimes struggles with likelihood minima and shapes, optimisations that target its robustness could make it more sensitive than both the optimal-observable estimator and the histogram method. These results underscore the potential of advanced simulation-based inference techniques, encouraging further exploration with LHC Run 3 data to surpass current ATLAS and CMS sensitivities.

Searching for HWW Anomalous Couplings with Simulation-Based Inference

Abstract

Understanding the source of the universe's asymmetry between matter and antimatter is one of the major open questions in particle physics. In this work, the sensitivity of novel machine-learning-based inference techniques to CP-odd and CP-even anomalous couplings is studied in the channel (), within the Standard Model Effective Field Theory (SMEFT) framework. Two machine-learning simulation-based inference (SBI) methods are explored: a per-event likelihood-ratio estimator, which directly approximates the ratio of probability densities between competing hypotheses, is benchmarked against a per-event optimal-observable estimator optimized for sensitivity to the parameters of interest. Both approaches are also compared to traditional summary statistics, in this case histograms of kinematic and angular observables, as commonly used in experimental analyses. SBI methods provide tighter constraints than one-dimensional summary statistics, though their performance is comparable to two-dimensional histogram analysis. The optimal-observable approach remains promising for its ability to probe multiple couplings simultaneously. Restricting the analysis to a region of high also enhances sensitivity to CP-odd operators while preserving sensitivity to CP-even operators, which histogram analyses often lose. Although the likelihood-ratio estimator sometimes struggles with likelihood minima and shapes, optimisations that target its robustness could make it more sensitive than both the optimal-observable estimator and the histogram method. These results underscore the potential of advanced simulation-based inference techniques, encouraging further exploration with LHC Run 3 data to surpass current ATLAS and CMS sensitivities.

Paper Structure

This paper contains 18 sections, 14 equations, 10 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Simulation-based Inference
  3. The likelihood-free inference problem
  4. Aiding the inference with augmented data
  5. Constraining $HWW$ anomalous couplings
  6. Current state-of-the-art
  7. Overview of the $WH \rightarrow \ell \nu b \bar{b}$ channel
  8. Energy-dependent and angular observables
  9. EFT scenarios studied
  10. Monte Carlo samples and event selection
  11. Sample unweighting and augmentation
  12. Training settings
  13. Limit setting
  14. Results
  15. Validation at parton-level
  16. ...and 3 more sections

Figures (10)

  • Figure 1: Feynman diagram illustrating the LO $WH$ associated production in the $l \nu b \bar{b}$ final state. The green circle corresponds to the vertex of interest.
  • Figure 2: Leading order Feynman diagram illustrating the most relevant backgrounds.
  • Figure 3: Distributions of the transverse mass of the ${\ell \nu b \bar{b}}$ system for the SM signal plus backgrounds (blue), compared to alternative BSM signal samples plus backgrounds, with $c_{HW} = 0.7$ (dashed green) and $c_{HW} = -0.7$ (dotted pink), obtained using the morphing technique. The lower panel shows the ratio to the SM prediction. The samples and selection cuts are described in Subsection \ref{['subsec:MC']}.
  • Figure 4: Distributions of the angular variable $Q_\ell \cos \delta^+$ for the SM signal plus backgrounds (blue), compared to alternative BSM signal samples plus backgrounds, with $c_{H\tilde{W}},c_{HW} =(0.7,-0.7)$ (dashed green) and $c_{H\tilde{W}},c_{HW} =(-0.7,-0.7)$ (dotted pink), obtained using the morphing technique. The lower panel shows the ratio to the SM prediction. The samples and selection cuts are described in Subsection \ref{['subsec:MC']}.
  • Figure 5: Estimated vs true log-likelihood ratio/score for the ALICES (left), ALICE (middle), and SALLY (right) techniques.
  • ...and 5 more figures