Lightweight Deep Learning Framework for Accurate Particle Flow Energy Reconstruction

2024-10-08

AI Summary

To address the low resolution, poor efficiency, and insufficient accuracy of classical particle-flow algorithms in high-multiplicity and heavily overlapping shower scenarios, this paper proposes a lightweight deep learning framework for high-precision particle-flow energy reconstruction. Methodologically, we design a hybrid loss function combining weighted mean squared error (MSE) and structural similarity (SSIM); integrate 3D convolutional layers with Squeeze-and-Excitation channel attention and offset-based self-attention modules to enhance cross-modal spatiotemporal modeling and nonlinear energy–displacement mapping; and incorporate multi-channel sparse feature representation. Experiments demonstrate that a lightweight model with only 90K parameters achieves performance comparable to a 5M-parameter baseline, while a 25M-parameter 3D model sets new state-of-the-art results on both interpolation and extrapolation tasks. All code and data preprocessing scripts are publicly released.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Under extreme operating conditions, characterized by high particle multiplicity and heavily overlapping shower energy deposits, classical particle flow algorithms encounter pronounced limitations in resolution, efficiency, and accuracy. To address this challenge, this paper proposes and systematically evaluates a deep learning reconstruction framework: For multichannel sparse features, we design a hybrid loss function combining weighted mean squared error with structural similarity index, effectively balancing pixel-level accuracy and structural fidelity. By integrating 3D convolutions, Squeeze-and-Excitation channel attention, and Offset self-attention modules into baseline convolutional neural networks, we enhance the model's capability to capture cross-modal spatiotemporal correlations and energy-displacement nonlinearities. Validated on custom-constructed simulation data and Pythia jet datasets, the framework's 90K-parameter lightweight variant approaches the performance of 5M-parameter baselines, while the 25M-parameter 3D model achieves state-of-the-art results in both interpolation and extrapolation tasks. Comprehensive experiments quantitatively evaluate component contributions and provide performance-parameter trade-off guidelines. All core code and data processing scripts are open-sourced on a GitHub repository to facilitate community reproducibility and extension.

Problem

• Improving particle flow energy reconstruction accuracy under extreme conditions
• Enhancing spatiotemporal correlation capture in deep learning models
• Balancing model performance and parameter efficiency for lightweight deployment

Innovation

• Hybrid loss function balances accuracy and fidelity
• 3D CNN with attention modules captures correlations
• Lightweight model matches performance of larger baselines