Learnable cut flow for high energy physics

Abstract

Neural networks have emerged as a powerful paradigm for tasks in high energy physics, yet their opaque training process renders them as a black box. In contrast, the traditional cut flow method offers simplicity and interpretability but requires extensive manual tuning to identify optimal cut boundaries. To merge the strengths of both approaches, we propose the Learnable Cut Flow (LCF), a neural network that transforms the traditional cut selection into a fully differentiable, data-driven process. LCF implements two cut strategies-parallel, where observable distributions are treated independently, and sequential, where prior cuts shape subsequent ones-to flexibly determine optimal boundaries. Building on this strategy, we introduce the Learnable Importance, a metric that quantifies feature importance and adjusts their contributions to the loss accordingly, offering model-driven insights unlike ad-hoc metrics. To ensure differentiability, a modified loss function replaces hard cuts with mask operations, preserving data shape throughout the training process. LCF is tested on six varied mock datasets and a realistic diboson vs. QCD dataset. Results demonstrate that LCF 1. accurately learns cut boundaries across typical feature distributions in both parallel and sequential strategies, 2. assigns higher importance to discriminative features with minimal overlap, 3. handles redundant or correlated features robustly, and 4. performs effectively in real-world scenarios. In the diboson dataset, LCF initially underperforms boosted decision trees and multiplayer perceptrons when using all observables. LCF bridges the gap between traditional cut flow method and modern black-box neural networks, delivering actionable insights into the training process and feature importance. Source code and experimental data are available at https://github.com/Star9daisy/learnable-cut-flow.

arXiv comment31 pages, 26 figures, 8 tables, source code and data are available at GitHub

References in corpus (24)

• Identifying Boosted Objects with N-subjettiness
• Jet Trimming
• ParticleNet: Jet Tagging via Particle Clouds
• Energy Correlation Functions for Jet Substructure
• Jet-Images -- Deep Learning Edition
• Jet Substructure Classification in High-Energy Physics with Deep Neural Networks
• Energy Flow Networks: Deep Sets for Particle Jets
• How Much Information is in a Jet?
• Novel Jet Observables from Machine Learning
• Mapping Machine-Learned Physics into a Human-Readable Space
• EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets
• Equivariant Energy Flow Networks for Jet Tagging
• Boosted $W/Z$ Tagging with Jet Charge and Deep Learning
• Automating the Construction of Jet Observables with Machine Learning
• Lorentz group equivariant autoencoders
• Model-agnostic search for dijet resonances with anomalous jet substructure in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• Pile-Up Mitigation using Attention
• GAN-AE : An anomaly detection algorithm for New Physics search in LHC data
• CALPAGAN: Calorimetry for Particles using GANs
• Full Event Particle-Level Unfolding with Variable-Length Latent Variational Diffusion
• Semi-supervised Graph Neural Networks for Pileup Noise Removal
• Searching for dark jets with displaced vertices using weakly supervised machine learning
• Applications of Lattice Gauge Equivariant Neural Networks
• Comparison of Machine Learning Approach to other Unfolding Methods