Quantum Annealing for Jet Clustering with Thrust

TL;DR

This paper explores the use of quantum annealing for jet clustering in high-energy physics, benchmarking its performance against classical methods and highlighting the importance of parameter tuning for optimal results.

Contribution

It demonstrates that quantum annealing can match classical approaches in jet clustering tasks when properly tuned, and introduces hybrid quantum/classical strategies.

Findings

Quantum annealing performs comparably to classical methods after tuning.

Hybrid quantum/classical approaches can achieve similar results without tuning.

Parameter tuning is crucial for quantum annealing effectiveness.

Abstract

Quantum computing holds the promise of substantially speeding up computationally expensive tasks, such as solving optimization problems over a large number of elements. In high-energy collider physics, quantum-assisted algorithms might accelerate the clustering of particles into jets. In this study, we benchmark quantum annealing strategies for jet clustering based on optimizing a quantity called "thrust" in electron-positron collision events. We find that quantum annealing yields similar performance to exact classical approaches and classical heuristics, but only after tuning the annealing parameters. Without tuning, comparable performance can be obtained through a hybrid quantum/classical approach.