Comparison of Image Processing Models in Quark Gluon Jet Classification

arXiv:2602.00141v1h-index: 6
Originality Synthesis-oriented
AI Analysis

This work addresses jet classification in high-energy physics, with incremental improvements in model efficiency and accuracy for domain-specific applications.

The paper tackled the problem of distinguishing quark and gluon jets using simulated jet images by comparing convolutional and transformer-based models, finding that fine-tuning only the final two blocks of the Swin-Tiny model achieved 81.4% accuracy and an AUC of 88.9%.

We present a comprehensive comparison of convolutional and transformer-based models for distinguishing quark and gluon jets using simulated jet images from Pythia 8. By encoding jet substructure into a three-channel representation of particle kinematics, we evaluate the performance of convolutional neural networks (CNNs), Vision Transformers (ViTs), and Swin Transformers (Swin-Tiny) under both supervised and self-supervised learning setups. Our results show that fine-tuning only the final two transformer blocks of the Swin-Tiny model achieves the best trade-off between efficiency and accuracy, reaching 81.4% accuracy and an AUC (area under the ROC curve) of 88.9%. Self-supervised pretraining with Momentum Contrast (MoCo) further enhances feature robustness and reduces the number of trainable parameters. These findings highlight the potential of hierarchical attention-based models for jet substructure studies and for domain transfer to real collision data.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes