Many Wrongs Make a Right: Leveraging Biased Simulations Toward Unbiased Parameter Inference

In particle physics, as in many areas of science, parameter inference relies on simulations to bridge the gap between theory and experiment. Recent developments in simulation-based inference have boosted the sensitivity of analyses; however, biases induced by simulation-data mismodeling can be difficult to control within standard inference pipelines. In this work, we propose a template-adapted mixture model to confront this problem in the context of signal fraction estimation: inferring the population proportion of signal in a mixed sample of signal and background, both of which follow arbitrarily complex distributions. We harness many biased simulations to perform data-driven estimates of each process distribution in the signal region, substantially reducing the bias on the signal fraction due to the domain shift between simulation and reality. We explore different methodological choices, including model selection, feature representation, and statistical method, and apply them to a Gaussian toy example and to a semirealistic di-Higgs measurement. We find that the presented methods successfully leverage the biased simulations to provide estimates with well-calibrated uncertainties.

Authors

Institutions

Publication Details

Journal
PRX Intelligence
Published
2026-11-18
DOI
https://doi.org/10.1103/lts9-9799
Citations
2
Primary Topic
Particle physics theoretical and experimental studies
Type
article
Field-Weighted Citation Impact
1.24

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Many Wrongs Make a Right: Leveraging Biased Simulations Toward Unbiased Parameter Inference

Sean Benevedes, Jesse Thaler, Ezequiel Álvarez, Manuel Szewc
2 citations
PRX Intelligence
Particle physics theoretical and experimental studies
1.24
article

Many Wrongs Make a Right: Leveraging Biased Simulations Toward Unbiased Parameter Inference

Sean Benevedes, Jesse Thaler, Ezequiel Álvarez, Manuel Szewc
article en
2 citations

Abstract

In particle physics, as in many areas of science, parameter inference relies on simulations to bridge the gap between theory and experiment. Recent developments in simulation-based inference have boosted the sensitivity of analyses; however, biases induced by simulation-data mismodeling can be difficult to control within standard inference pipelines. In this work, we propose a template-adapted mixture model to confront this problem in the context of signal fraction estimation: inferring the population proportion of signal in a mixed sample of signal and background, both of which follow arbitrarily complex distributions. We harness many biased simulations to perform data-driven estimates of each process distribution in the signal region, substantially reducing the bias on the signal fraction due to the domain shift between simulation and reality. We explore different methodological choices, including model selection, feature representation, and statistical method, and apply them to a Gaussian toy example and to a semirealistic di-Higgs measurement. We find that the presented methods successfully leverage the biased simulations to provide estimates with well-calibrated uncertainties.

PRX IntelligenceVol. 1(2)
Consejo Nacional de Investigaciones Científicas y Técnicas (AR), National University of General San Martín (AR), Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR), CEA Paris-Saclay (FR), International Center for Advanced Studies (AR), The NSF AI Institute for Artificial Intelligence and Fundamental Interactions (US), Institut de Physique Théorique (FR), Massachusetts Institute of Technology (US), Institut des Hautes Études Scientifiques (FR)
National Science Foundation, U.S. Department of Energy, Simons Foundation, Institut des Hautes Études Scientifiques, Université Paris-Saclay, High Energy Physics
Openalex Percentile: Top 24%
Particle physics theoretical and experimental studies
1.24
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon