Quantum Generative Adversarial Networks For Anomaly Detection In High Energy Physics

Abstract

The standard model (SM) of particle physics represents a theoretical paradigm for the description of the fundamental forces of nature. Despite its broad applicability, the SM does not enable the description of all physically possible events. The detection of events that cannot be described by the SM, which are typically referred to as anomalous, and the related potential discovery of exotic physical phenomena is a non-trivial task. The challenge becomes even greater with next-generation colliders that will produce even more events with additional levels of complexity. The additional data complexity motivates the search for unsupervised anomaly detection methods that do not require prior knowledge about the underlying models. In this work, we develop such a technique. More explicitly, we employ a quantum generative adversarial network to identify anomalous events. The method learns the background distribution from SM data and, then, determines whether a given event is characteristic for the learned background distribution. The proposed quantum-powered anomaly detection strategy is tested on proof-of-principle examples using numerical simulations and IBM Quantum processors. We find that the quantum generative techniques using ten times fewer training data samples can yield comparable accuracy to the classical counterpart for the detection of the Graviton and Higgs particles. Additionally, we empirically compute the capacity of the quantum model and observe an improved expressivity compared to its classical counterpart.

arXiv commentFor updated final version, see proceedings of the 2023 IEEE International Conference on Quantum Computing and Engineering (QCE), vol 01, pages 331-341

References in corpus (35)

• Adam: A Method for Stochastic Optimization
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Barren plateaus in quantum neural network training landscapes
• The theory of variational hybrid quantum-classical algorithms
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Quantum Circuit Learning
• On the Convergence of Adam and Beyond
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Searching for Exotic Particles in High-Energy Physics with Deep Learning
• Parameterized quantum circuits as machine learning models
• Cost Function Dependent Barren Plateaus in Shallow Parametrized Quantum Circuits
• Evaluating analytic gradients on quantum hardware
• Quantum generative adversarial learning
• Quantum generative adversarial networks
• General-purpose event generators for LHC physics
• Quantum Generative Adversarial Networks for Learning and Loading Random Distributions
• Hierarchical quantum classifiers
• Quantum gradient descent for linear systems and least squares
• Dijet resonance search with weak supervision using $\sqrt{s}=13$ TeV $pp$ collisions in the ATLAS detector
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider
• JUNIPR: a Framework for Unsupervised Machine Learning in Particle Physics
• Generating and designing DNA with deep generative models
• Quantum Machine Learning in High Energy Physics
• Event Classification with Quantum Machine Learning in High-Energy Physics
• Challenges for Unsupervised Anomaly Detection in Particle Physics
• Interpretable Deep Learning for Two-Prong Jet Classification with Jet Spectra
• Variational learning for quantum artificial neural networks
• Quantum anomaly detection in the latent space of proton collision events at the LHC
• Unravelling physics beyond the standard model with classical and quantum anomaly detection
• MUSiC: a model-unspecific search for new physics in proton-proton collisions at $\sqrt{s}=$ 13 TeV
• Particle Track Reconstruction with Quantum Algorithms
• Cost function embedding and dataset encoding for machine learning with parameterized quantum circuits
• Deep Learning Jet Image as a Probe of Light Higgsino Dark Matter at the LHC
• A scale-dependent notion of effective dimension

Cited by in corpus (6)

• Quantum Quantile Mechanics: Solving Stochastic Differential Equations for Generating Time-Series
• Exploring Unsupervised Anomaly Detection with Quantum Boltzmann Machines in Fraud Detection
• Loop Feynman integration on a quantum computer
• Dead Gate Elimination
• Flowing Through Hilbert Space: Quantum-Enhanced Generative Models for Lattice Field Theory
• Quantum Chebyshev Probabilistic Models for Fragmentation Functions