Quantum Rationale-Aware Graph Contrastive Learning for Jet Discrimination

TL;DR

This paper introduces a quantum rationale-aware graph contrastive learning framework that enhances jet discrimination in high-energy physics, achieving competitive results with fewer parameters and limited labeled data.

Contribution

The study develops a novel quantum rationale generator integrated into a graph contrastive learning framework for resource-efficient jet discrimination.

Findings

Achieves 77.5% AUC on quark-gluon jet dataset.

Uses only 45 quantum parameters, demonstrating efficiency.

Outperforms classical and hybrid benchmarks.

Abstract

In high-energy physics, particle jet tagging plays a pivotal role in distinguishing quark from gluon jets using data from collider experiments. While graph-based deep learning methods have advanced this task beyond traditional feature-engineered approaches, the complex data structure and limited labeled samples present ongoing challenges. More broadly, our primary focus is the development of a rationale-aware graph contrastive learning framework designed to operate under strict resource constraints; we use quark-gluon jet discrimination as a representative and practically relevant use case. However, existing contrastive learning (CL) frameworks struggle to leverage rationale-aware augmentations effectively, often lacking supervision signals that guide the extraction of salient features and facing computational efficiency issues such as high parameter counts. In this study, we demonstrate that integrating a quantum rationale generator (QRG) within our proposed Quantum Rationale-aware Graph Contrastive Learning (QRGCL) framework enables competitive jet discrimination performance, particularly in parameter-constrained settings, reducing reliance on labeled data, and capturing rationale-aware features. Evaluated on the quark-gluon jet dataset, QRGCL achieves an AUC score of $77.5\%$ while maintaining a compact architecture of only 45 QRG parameters, achieving competitive performance compared to classical, quantum, and hybrid benchmarks. These results highlight QRGCL's potential to advance jet tagging and other complex classification tasks in high-energy physics, where computational efficiency and limitations in feature extraction persist.