Quantum Support Vector Machines for Continuum Suppression in B Meson Decays

Abstract

Quantum computers have the potential to speed up certain computational tasks. A possibility this opens up within the field of machine learning is the use of quantum techniques that may be inefficient to simulate classically but could provide superior performance in some tasks. Machine learning algorithms are ubiquitous in particle physics and as advances are made in quantum machine learning technology there may be a similar adoption of these quantum techniques. In this work a quantum support vector machine (QSVM) is implemented for signal-background classification. We investigate the effect of different quantum encoding circuits, the process that transforms classical data into a quantum state, on the final classification performance. We show an encoding approach that achieves an average Area Under Receiver Operating Characteristic Curve (AUC) of 0.848 determined using quantum circuit simulations. For this same dataset the best classical method tested, a classical Support Vector Machine (SVM) using the Radial Basis Function (RBF) Kernel achieved an AUC of 0.793. Using a reduced version of the dataset we then ran the algorithm on the IBM Quantum ibmq_casablanca device achieving an average AUC of 0.703. As further improvements to the error rates and availability of quantum computers materialise, they could form a new approach for data analysis in high energy physics.

Quantum Computing for Data Analysis in High-Energy Physics

“…Another implementation of a QSVM algorithm for classification is described in Ref. [68]. Here, the authors designed and implemented a QSVM approach for the signal-background classification task in B meson decays.…”

Snowmass White Paper: Quantum Computing Systems and Software for High-energy Physics Research

“…One example includes algorithms to construct physics objects amenable to analysis from the signals generated in a particle detector-i.e., the clustering of detector hits into so-called tracks for reconstructing a particle's trajectory [81][82][83][84][85][86][87][88] or tracks and calorimeter energy depositions into jets [89][90][91]. Furthermore, quantum-assisted algorithms have been explored in unsupervised learning settings to classify jets according to their origin (b-tagging) [92], generative tasks [93][94][95], and the selection of events or interactions along with background suppression [3,13,[96][97][98][99][100][101][102][103][104][105][106][107][108]. In particular, generative models have been explored extensively as an alternative for the simulation of particle interactions and the detector's response to such interactions [4,94].…”

Unravelling physics beyond the standard model with classical and quantum anomaly detection

“…Quantum Machine Learning (QML) has recently been proposed as a new framework offering potential speedups and performance improvements over classical ML [16][17][18][19]. Several QML algorithms, like Quantum Support Vector Classifiers (QSVCs) [20][21][22], Variational Quantum Classifiers (VQCs) [21,22], Quantum Convolutional Neural Networks (QCNNs) [23], or quantum autoencoders [24] have been applied to a wide range of HEP problems [25][26][27][28][29][30][31][32][33][34]. With the current methods, quantum algorithms generally achieve a performance similar to their classical counterparts.…”