Abstract: This study addresses the "LHC inverse problem"—the challenge of identifying a unique theoretical model from experimental data. By simulating over 40,000 Supersymmetry (SUSY) models, the authors demonstrate that the mapping from signature space to parameter space is highly degenerate, with multiple distinct theories often producing identical collider signatures.
Introduction
In the field of high-energy physics, the "forward problem" is a well-established process: starting from a specific theoretical framework, such as Supersymmetry (SUSY), physicists calculate parameters and predict the resulting experimental signatures at particle colliders. However, the "inverse problem" is significantly more daunting. If the Large Hadron Collider (LHC) detects a surplus of events—such as jets accompanied by missing transverse energy—how can researchers determine which specific theoretical model is responsible?
The primary challenge stems from the high dimensionality of the theory. The Minimal Supersymmetric Standard Model (MSSM) contains over 100 parameters. Even simplified versions typically involve at least 15. Because detectors are imperfect and certain particles (like dark matter candidates) escape undetected, distinct sets of physical parameters can "masquerade" as one another, producing identical experimental data. This paper investigates this mapping, identifying "degeneracies" where multiple theories lead to the same signature.
Methodology
The authors conducted a large-scale simulation study to map the relationship between the theoretical Parameter Space and the experimental Signature Space.
- Model Generation: They generated 43,026 distinct MSSM models using a 15-dimensional parameterization, covering gaugino masses, squark/slepton masses, and $\tan \beta$.
- Simulation Pipeline: For each model, 10 fb⁻¹ of LHC data was simulated using PYTHIA for event generation and PGS for detector simulation.
- Observables: The study tracked 1,808 different observables, including event counts, lepton multiplicities, and complex kinematic distributions (deciles and 20-tiles of invariant masses).
- Statistical Analysis: To estimate the number of "distinguishable" models, the authors employed a technique inspired by the "birthday paradox." By measuring how often randomly sampled models produced identical signatures, they could estimate the total number of unique "bins" in the signature space.
Theoretical Insights: The Nature of Degeneracy
The study reveals that the inverse map is not a continuous blur but is instead "patchy," consisting of isolated "islands" in parameter space. A single experimental signature does not necessarily point to a unique theoretical origin; rather, it often points to several disconnected regions.
The authors identified three primary mechanisms for these degeneracies:
- Flippers: Models that swap particle identities (e.g., a Wino-like vs. a Higgsino-like Lightest Supersymmetric Particle) while maintaining similar mass scales.
- Sliders: Models where the entire mass spectrum is shifted up or down, but the mass differences—to which colliders are most sensitive—remain constant.
- Squeezers: Models that "hide" a particle by minimizing the mass gap between it and the next state, causing its decay products to be too "soft" for detector triggers.
Furthermore, despite the use of 1,808 observables, the "effective dimensionality" of the signature space was found to be only 5 or 6. This high degree of correlation explains why different models can so easily mimic one another.
Key Results
The simulation results quantify the scale of the inverse problem:
- Degeneracy Scale: For a typical experimental signature, there are approximately 10 to 100 qualitatively different models that could explain the data.
- Distinguishable Outcomes: The authors estimate there are roughly $3 \times 10^6$ experimentally distinguishable outcomes at the LHC within the MSSM framework.
- The Power of Leptons: The presence of light "sleptons" (superpartners of electrons and muons) significantly reduces degeneracy. Leptons provide more information (flavor and charge) than hadronic jets, making it much harder for models to masquerade as one another.
Conclusion and Future Directions
While this study provides a framework for understanding the LHC inverse problem, several factors may complicate the real-world application. The exclusion of Standard Model backgrounds (such as $t\bar{t}$ or $W/Z$ + jets) likely underestimates the number of degeneracies, as background noise can "wash out" subtle signature differences.
Future work will need to address higher luminosity data (e.g., 300 fb⁻¹), which may break some degeneracies through increased precision. However, discrete ambiguities like "Flippers" may persist. The authors emphasize the necessity of developing "orthogonal" signatures—novel ways of analyzing data that are uncorrelated with existing observables—to uniquely pinpoint the underlying physical laws.
Summary updated