Pileup Per Particle Identification

Abstract: We propose a new method for pileup mitigation by implementing "pileup per particle identification" (PUPPI). For each particle we first define a local shape $\alpha$ which probes the collinear versus soft diffuse structure in the neighborhood of the particle. The former is indicative of particles originating from the hard scatter and the latter of particles originating from pileup interactions. The distribution of $\alpha$ for charged pileup, assumed as a proxy for all pileup, is used on an event-by-event basis to calculate a weight for each particle. The weights describe the degree to which particles are pileup-like and are used to rescale their four-momenta, superseding the need for jet-based corrections. Furthermore, the algorithm flexibly allows combination with other, possibly experimental, probabilistic information associated with particles such as vertexing and timing performance. We demonstrate the algorithm improves over existing methods by looking at jet $p_T$ and jet mass. We also find an improvement on non-jet quantities like missing transverse energy.

• The paper introduces a novel particle-level method using a local shape variable to distinguish primary interactions from pileup effects.
• It employs event-by-event charged particle characterization to assign weights, substantially improving jet pT, mass resolution, and MET measurements.
• The technique’s adaptability with vertexing and timing data offers promising potential for future high-energy collider experiments.

Overview of "Pileup Per Particle Identification" Paper

The article "Pileup Per Particle Identification" by Bertolini et al. introduces a method dubbed PileUp Per Particle Identification (PUPPI) for mitigating the effects of pileup in high luminosity hadron collider environments, such as those expected in future LHC runs. The authors propose an innovative approach to identify and mitigate pileup effects at the particle level, thereby improving the treatment of pileup in complex environments, particularly for collider experiments.

Methodology

The PUPPI algorithm works by defining a local shape variable, denoted as $\alpha$, for each particle in the event. The local shape aims to distinguish tracks emanating from hard scattering—the primary interaction—from those associated with pileup events. The variable $\alpha$ differentiates based on local collinear patterns typical of a parton shower versus the unstructured nature of pileup interactions.

To implement pileup mitigation, an event-by-event characterization of pileup distributions is performed using only charged pileup particles, exploiting efficient tracking information within the central region of the detector. Subsequently, each particle is assigned a weight that reflects its likelihood of contributing to the pileup. These weights are calculated in comparison to the average pileup shape determined from charged particles. The algorithm then reweights individual particles' momenta using these assigned weights to mitigate pileup influence comprehensively. Moreover, the approach is robust to incorporation of experimental probabilistic information, such as vertexing and calorimeter timing, to refine the weight calculation further.

Results

The empirical evaluation demonstrates that the PUPPI algorithm substantially improves pileup corrections compared to existing strategies such as traditional four-vector subtraction and charged hadron subtraction (CHS). When applied in simulation studies, PUPPI shows marked improvements in jet $p_T$ and mass resolution across various jet kinematic properties and pileup severity levels. Importantly, PUPPI also enhances non-jet related observables such as missing transverse energy (MET), presenting an efficient methodology for relieving pileup effects in the calculation and deduction of MET from collider data.

Implications and Future Directions

The PUPPI method offers a consistent event interpretation by operating at the particle level, thereby circumventing deficiencies associated with jet-based pileup mitigation alone. Unlike global event correction methods, PUPPI accounts for local fluctuations and allows a more detailed correction, particularly beneficial in challenging high-pileup scenarios. The particle-centric strategy is scalable and adaptable to incorporate non-particle information, reflecting its potential for deployment in practical experimental setups.

The implications for scaling to future experiments are significant, where increasing pileup presents daunting challenges. PUPPI's ability to function effectively with minimal dependence on tracking data extends its utility into forward regions, potentially influencing future detector designs to exploit such flexible algorithms for pileup mitigation.

Conclusion

The PileUp Per Particle Identification method provides a promising and effective approach to managing pileup contamination in high-energy particle physics experiments. Its adaptability, flexibility, and locality of corrections present an intriguing advancement over existing methodologies. As experimental conditions at colliders continue to evolve, techniques such as PUPPI will be invaluable in ensuring the precise reconstruction of physics events and maintaining the sensitivity and accuracy of particle physics measurements. Further research may explore optimizing $\alpha$ calculations and refining its adaptability across a broader range of experimental conditions to enhance performance across diverse physics analyses.