Invertible Neural Networks (INN) have become established tools for the simulation and generation of highly complex data. We propose a quantum-gate algorithm for a Quantum Invertible Neural Network (QINN) and apply it to the LHC data of jet-associated production of a Z-boson that decays into leptons, a standard candle process for particle collider precision measurements. We compare the QINN’s performance for different loss functions and training scenarios. For this task, we find that a hybrid QINN matches the performance of a significantly larger purely classical INN in learning and generating complex data.
Invertible Neural Networks (INN) have become established tools for the simulation and generation of highly complex data. We propose a quantum-gate algorithm for a Quantum Invertible Neural Network (QINN) and apply it to the LHC data of jet-associated production of a Z-boson that decays into leptons, a standard candle process for particle collider precision measurements. We compare the QINN’s performance for different loss functions and training scenarios. For this task, we find that a hybrid QINN matches the performance of a significantly larger purely classical INN in learning and generating complex data.
The fast calorimeter simulation is a critical computational bottleneck for the ATLAS experiment atthe High-Luminosity LHC and for the design of future colliders. Invertible Neural Networks (INNs) andnormalizingflowshavedemonstratedimpressiveaccuracyasgenerativemodelsforcalorimeter shower simulation, forming the backbone of the emerging FastCaloSim framework. In this work we explore the extension of this classical approach toward quantum-conditioned invertible neural networks (qINN). We present the conceptual motivation, discuss the fundamental tension between the unitarity of quantum circuits and the irreversibility of projective measurement, and out line two complementary solution strategies. As a first feasibility study we implement a hybrid architectureinwhichasmallquantumcircuitreplacestheclassicalconditioningsub-network inside an otherwise classicalINN, trained on ATLAS Open-Data pion showers from the FastCalorimeter Challenge dataset. The preliminary results, obtained with a 6-qubit, 2-layer variational ansatz, show encouraging convergence and motivate further optimisation. The work is a feasibility study; no definitive performance claims are made.
Classical deep neural networks can learn rich multi-particle correlations in collider data, but their inductive biases are rarely anchored in physics structure. We propose quantum-informed neural networks (QINNs), a general framework that brings quantum information concepts and quantum observables into purely classical models. While the framework is broad, in this paper, we study one concrete realisation that encodes each particle as a qubit and uses the Quantum Fisher Information Matrix (QFIM) as a compact, basis-independent summary of particle correlations. Using jet tagging as a case study, QFIMs act as lightweight embeddings in graph neural networks, increasing model expressivity and plasticity. The QFIM reveals distinct patterns for QCD and hadronic top jets that align with physical expectations. Thus, QINNs offer a practical, interpretable, and scalable route to quantum-informed analyses, that is, tomography, of particle collisions, particularly by enhancing well-established deep learning approaches.
Jets are central to the physics programs of both current and future colliders, from precision Standard Model measurements and searches for new physics at the Large Hadron Collider to studies of nucleon structure at the future Electron-Ion Collider. Motivated by these applications, we explore quantum machine learning for jet classification and present a permutation-invariant Quantum Graph Neural Network (QGNN) applied to particle-cloud representations of jets. We apply the model to two such discrimination tasks: quark vs. gluon and up vs. down quark flavor tagging, with the latter being, to our knowledge, the first application of a quantum model to this problem. In the ideal simulation, the QGNN performs competitively against the Particle Flow Network and traditional QCD observables. We further deploy scaled-down models to IBM and IonQ quantum processing units (QPUs), where we train and evaluate them, obtaining promising results. Finally, we perform an interpretability analysis to characterize the observables learned by the quantum model, relating them to generalized angularities for the quark-gluon study and to jet charge for the flavor study.
Two strong transformative fields in contemporary artificial intelligence (AI) are quantum computing and deep learning. Neuro-nanophotonic systems also introduce quantum-inspired innovations that enhance the computing power of neural networks. Quantum-Inspired Neural Networks (QINNs) have shown promise as a potential alternative to the classical deep learning models, particularly when processing high-dimensional data and solving complex problems efficiently. QINNs leverage the attributes of quantum mechanics to enable efficient deep learning methodologies, achieving superior performance in AI-based applications. In this research, we present a QINN framework that synergizes the power of Quantum Variational Circuits (QVCs) and Quantum Natural Gradient Descent (QNGD) to improve learning efficiency of learning. Experimental results show that QINNs can achieve 97% accuracy, which is 4.5% improvement over the classical models, and computational time and energy consumption can be reduced by 45% and 53%, respectively. Furthermore, QINNs exhibited greater robustness to adversarial noise, making them better candidates for healthcare, cybersecurity, and large-scale AI applications. The results illuminate the game-changing ability of QINNs with respect to scalable, cross-computational and highly resilient AI architectures.
Machine Learning algorithms are playing a fundamental role in solving High Energy Physics tasks. In particular, the classification of hadronic jets at the Large Hadron Collider is suited for such types of algorithms, and despite the great effort that has been put in place to tackle such a classification task, there is room for improvement. In this context, Quantum Machine Learning is a new methodology that takes advantage of the intrinsic properties of quantum computation (e.g. entanglement between qubits) to possibly improve the performance of a classification task. In this contribution, a new study of Quantum Machine Learning applied to jet identification is presented. Namely, a Variational Quantum Classifier is trained and evaluated on fully simulated data of the LHCb experiment, in order to identify jets containing a hadron formed by a b or bˉ quark at the moment of production. The jet identification performance of the quantum classifier is compared with a Deep Neural Network using the same input features.
Abstract In this study, we compare the performance of randomized classical and quantum neural networks (NNs) as well as classical and quantum-classical hybrid convolutional neural networks (CNNs) for the task of supervised binary image classification. We keep the employed quantum circuits compatible with near-term quantum devices and use two distinct methodologies: applying randomized NNs on dimensionality-reduced data, and applying CNNs to full image data. We evaluate these approaches on three fully-classical data sets of increasing complexity: an artificial hypercube data set, MNIST handwritten digits and industrial images of practical relevance. Our study’s central goal is to shed more light on how quantum and classical models perform for various binary classification tasks and on what defines a good quantum model. To this end, our study involves a correlation analysis between classification accuracy and quantum model hyperparameters, and an analysis on the role of entanglement in quantum models, as well as on the impact of initial training parameters. We find classical and quantum-classical hybrid models achieve statistically-equivalent classification accuracies across most data sets with no approach consistently outperforming the other. Interestingly, we observe that quantum NNs show lower variance with respect to initial training parameters and that the role of entanglement is nuanced. While incorporating entangling gates seems to be generally advantageous, we also observe that their (optimizable) entangling power is not correlated with model performance. We also observe an inverse proportionality between the number of entangling gates and the average gate entangling power. Our study provides an industry perspective on quantum machine learning for practical binary image classification tasks, highlighting both current limitations and potential avenues for further research in quantum circuit design, entanglement utilization, and model transferability across varied applications.