# Pileup Mitigation with Machine Learning (PUMML)

arXiv: https://arxiv.org/abs/1707.08600 · HTML: https://mlsanity.com/paper/1707.08600/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jul 2017.

## Built on by (most cited first, 9 of 9)

- [A Living Review of Machine Learning for Particle Physics](https://mlsanity.com/paper/2102.02770/) (2021)
- [Structural Re-weighting Improves Graph Domain Adaptation](https://mlsanity.com/paper/2306.03221/) (2023)
- [Pairwise Alignment Improves Graph Domain Adaptation](https://mlsanity.com/paper/2403.01092/) (2024)
- [Bias and Priors in Machine Learning Calibrations for High Energy Physics](https://mlsanity.com/paper/2205.05084/) (2022)
- [The Radio-Frequency Transformer for Signal Separation](https://mlsanity.com/paper/2603.09201/) (2026)
- [Structural Alignment Improves Graph Test-Time Adaptation](https://mlsanity.com/paper/2502.18334/) (2025)
- [Variational inference for pile-up removal at hadron colliders with diffusion models](https://mlsanity.com/paper/2410.22074/) (2024)
- [Deep-learning-based decomposition of overlapping-sparse images: application at the vertex of neutrino interactions](https://mlsanity.com/paper/2310.19695/) (2023)
- [PhyGHT: Physics-Guided HyperGraph Transformer for Signal Purification at the HL-LHC](https://mlsanity.com/paper/2602.20475/) (2026)

## Abstract (arXiv)

Pileup involves the contamination of the energy distribution arising from the primary collision of interest (leading vertex) by radiation from soft collisions (pileup). We develop a new technique for removing this contamination using machine learning and convolutional neural networks. The network takes as input the energy distribution of charged leading vertex particles, charged pileup particles, and all neutral particles and outputs the energy distribution of particles coming from leading vertex alone. The PUMML algorithm performs remarkably well at eliminating pileup distortion on a wide range of simple and complex jet observables. We test the robustness of the algorithm in a number of ways and discuss how the network can be trained directly on data.