Variational Quantum Algorithms for Particle Track Reconstruction
Vincenzo Lipardi 1 †, Xenofon Chiotopoulos 1,2,3 †, Jacco A. de Vries 2,3, Domenica Dibenedetto 1, Kurt Driessens 1, Marcel Merk 2,3, Mark H.M. Winands 1 · 2025
Read Variational Quantum Algorithms for Particle Track Reconstruction by Vincenzo Lipardi 1 †, Xenofon Chiotopoulos 1,2,3 †, Jacco A. de Vries 2,3, Domenica Dibenedetto 1, Kurt Driessens 1, Marcel Merk 2,3, Mark H.M. Winands 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-11397.txt.
Introduction
Vincenzo Lipardi, Xenofon Chiotopoulos, and their colleagues test variational quantum algorithms on particle track reconstruction, the task of tracing particle paths through layered detectors in high-energy physics. Quantum methods promise help as computing demands grow, but their circuit designs must be both expressive and efficient for real event shapes. Two formulations of the tracking problem, modeled on a collider vertex detector, are compared.
One frames tracking as a ground-state energy problem; the other frames it as a system of linear equations. A Monte Carlo tree search designs the quantum circuits for each problem size. Experiments weigh performance against computing cost for both formulations across sizes.
Catalog record
| CREATOR | Vincenzo Lipardi 1 †, Xenofon Chiotopoulos 1,2,3 †, Jacco A. de Vries 2,3, Domenica Dibenedetto 1, Kurt Driessens 1, Marcel Merk 2,3, Mark H.M. Winands 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-11397
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Variational Quantum Algorithms for Particle Track Reconstruction Vincenzo Lipardi 1 †, Xenofon Chiotopoulos 1,2,3 †, Jacco A. de Vries 2,3, Domenica Dibenedetto 1, Kurt Driessens 1, Marcel Merk 2,3, Mark H.M. Winands 1 Abstract Quantum Computing is a rapidly developing field with the potential to tackle the increasing computational challenges faced in high-energy physics. In this work, we explore the potential and limitations of variational quantum algorithms in solving the particle track reconstruction problem. We present an analysis of two distinct formulations for identifying straight-line tracks in a multilayer detection system, inspired by the LHCb vertex detector. The first approach is formulated as a ground-state energy problem, while the second approach is formulated as a system of linear equations. This work addresses one of the main challenges when dealing with variational quantum algorithms on general problems, namely designing an expressive and efficient quantum ansatz working on tracking events with fixed detector geometry. For this purpose, we employed a quantum architecture search method based on Monte Carlo Tree Search to design the quantum circuits for different problem sizes. We provide experimental results to test our approach on both formulations for different problem sizes in terms of performance and computational cost. † † footnotetext: These authors contributed equally to this work. The order of the first two authors has been chosen by flipping a quantum coin. 1 Introduction In large-scale particle physics experiments, particles are collided at high energies and at high frequencies of 40 MHz, in order to study the fundamental forces of nature. In a single collision event, hundreds of new particles are simultaneously produced and traverse through sensitive detection layers where they deposit small amounts of energy, resulting in so-called hits in the detectors. These hits are then reconstructed with fast algorithms into particle trajectories or tracks that are used in subsequent analyses. In the upcoming High Luminosity phase of the Large Hadron Collider (LHC), the number of simultaneous collisions will increase significantly leading to the production of unprecedented data volumes to be processed. The increase in complexity presents a challenge still to be resolved, since the track reconstruction task scales to the power of 2-3 with the number of hits per layer. x Various approaches have been pursued [ 14 ] , at the time of writing the most performant approach is based on GPU parallel track reconstruction of events [ 7 ] . There have been attempts to leverage quantum computing to solve the tracking problem. As tracking can be expressed as a Quadratic Unconstrained Binary Optimization (QUBO) problem many quantum algorithms can be applied. Some popular approaches are; Quantum Graph Neural Networks [ 12 , 19 , 41 ] , quantum annealing [ 3 , 35 ] , quantum annealing-inspired algorithms [ 31 ] and Variational Quantum Eigensolver (VQE) using a sub-QUBO formulation of the problem and tested it on real hardware [ 18 ] . Moreover, the tracking problem has been mapped to a linear system of equations and in [ 28 ] the authors prove that it fulfills all the properties necessary to employ the Harrow–Hassidim–Lloyd (HHL) algorithm, which gives theoretical guarantees for an exponential advantage compared to classical techniques in terms of computational complexity. Although experimental results have shown promising results, a more extensive application of the HHL algorithm is still unfeasible on Near Intermediate-Scale Quantum (NISQ) devices [ 33 ] . x In this article, we employ the hardware-tailored approach of the Variational Quantum Algorithms (VQAs) for the particle tracking problem, that allows consideration of the user’s available hardware specifications. We address one of the main limitations of these methods, that is the choice of the quantum circuit ansatz [ 8 ] , as also highlighted for our specific problem in [ 18 ] . VQAs are hybrid quantum-classical algorithms where the problem of interest is encoded into an optimization task over parameterized quantum circuits. The choice of quantum circuit to optimize on, known as ansatz , significantly affects the performance of the algorithm [ 8 ] . In this context, it the automatic design of ansatz for VQA emerged as a relevant research direction, known as quantum architecture search , or also quantum ansatz search . It can be seen as the quantum analogue of designing the structure on Artificial Neural Networks [ 34 ] . For this scope, we employ an automated quantum ansatz search technique introduced in [ 26 ] based on Monte Carlo Tree Search (MCTS) [ 11 ] . Moreover, we provide a deeper overview on the performance of Variational Quantum Algorithms [ 8 ] for particle track reconstruction by considering two different formulations. One is a ground state energy problem and is solved through a VQE [ 23 ] , similar to [ 18 ] . While the other is encoded in a system of linear equations, equivalent to the HHL formulation [ 29 ] and solved through the Variational Quantum Linear Solver (VQLS) [ 5 ] . Note that the ansatz search problem relates to both formulations. Hence, MCTS is used to design the ansatz for the VQE and VQLS independently and for all the tested problem sizes. The goal is to find an ansatz that solves the tracking problem for different problem sizes and configurations. To test our approach we provide experimental results based on noiseless simulations of both problem formulations. 2 Particle Track Reconstruction In this paper, we address the particle tracking problem for a simplified version of the VELO subdetector of the LHCb experiment at the LHC at CERN [ 10 ] . This is the subdetector immediately surrounding the particle collision point. This detector consists of two sides of 26 vertically oriented pixel sensors arranged along the beam line positioned around the LHC collision point. The detector geometry, particle hits and particle tracks can be seen in Figure 1 . This figure shows the complexity of a simulated event under a realistic detector response. One key aspect of the VELO is that due to the lack of a magnetic field, the particle tracks are straight lines, which can also be seen in Figure 1 . Figure 1: A full event example of the LHCb VELO detector. The white dots represent the detector hits and the blue lines are the reconstructed tracks. Credit: D. Nicotra. Solving the particle track reconstruction using quantum algorithms requires us to find a representation of the problem in such a way that it takes advantage of the quantum computation. Considering the detector geometry we construct a fully connected graph where the graph layers are the detector modules. Each layer has all the hits that pass through that detector layer and each connection represents a possible track segment (doublet). Figure 2: Graph representation of an event with 5 detector layers and 6 particles (Toy Model). The vertical lines represent the detection layers, and the red circles are the hits interconnected with doublets/segments. The full-color segments are part of the found tracks, the grey segments are combinatoric backgrounds. Where S_{i} and S_{j} are doublets with their value indicating if they are part of a track [ 28 ] . We assign a binary value to each track segment, S_{i}=\{0,1\} , where a 1 indicates the doublet being part of