Di-Higgs to 4b with Bayesian inference: improving simulation estimates

TL;DR

This paper introduces a Bayesian mixture model that enhances di-Higgs to 4b analysis by directly inferring signal and background, correcting biases, and improving sensitivity despite simulation challenges.

Contribution

The novel Bayesian approach simultaneously infers signal and background shapes within the signal region, improving calibration and sensitivity over traditional methods.

Findings

Corrects biased priors and provides calibrated credible intervals.

Improves ROC/AUC metrics over cut-and-count baselines.

Demonstrates robustness with simulated mismatched data.

Abstract

Measuring di-Higgs production in the four-bottom channel is challenged by overwhelming QCD backgrounds and imperfect simulations. We develop a Bayesian mixture model that simultaneously infers signal and background fractions and their individual shapes directly in the signal region. The likelihood is a nuanced combination of a one-dimensional kinematic discriminator and per-jet flavour scores; with their correlations incorporated via kinematic bins. Monte Carlo informs weak Dirichlet priors, while the posterior adjusts to the interplay of the model, priors and observed data. Using pseudo-data simulated with standard tools and with controlled mismatches, we show that the method corrects biased priors, delivers calibrated 68-95% credible intervals for the signal count, and improves dataset-level ROC/AUC relative to simple cut-and-count baselines. This study highlights how Bayesian inference can harvest information present in the signal region and self-calibrate model parameters, providing a robust route to increased sensitivity in di-Higgs searches.