Quantum Neural Network Research Articles

Quantum Fuzzy Neural Network for multimodal sentiment and sarcasm detection

Sentiment and sarcasm detection in social media contribute to assessing social opinion trends. Over the years, most artificial intelligence (AI) methods have relied on real values to characterize the sentimental and sarcastic features in language. These methods often overlook the complexity and uncertainty of sentimental and sarcastic elements in human language. Therefore, this paper proposes the Quantum Fuzzy Neural Network (QFNN), a multimodal fusion and multitask learning algorithm with a Seq2Seq structure that combines Classical and Quantum Neural Networks (QNN), and fuzzy logic. Complex numbers are used in the Fuzzifier to capture sentiment and sarcasm features, and QNN are used in the Defuzzifier to obtain the prediction. The experiments are conducted on classical computers by constructing quantum circuits in a simulated noisy environment. The results show that QFNN can outperform several recent methods in sarcasm and sentiment detection task on two datasets (Mustard and Memotion). Moreover, by assessing the fidelity of quantum circuits in a noisy environment, QFNN was found to have excellent robustness. The QFNN circuit also possesses expressible and entanglement capabilities, proving effective in various settings. Our code is available at https://github.com/prayagtiwari/QFNN.

Hybrid framework for respiratory lung diseases detection based on classical CNN and quantum classifiers from chest X-rays

The human respiratory system might be seriously affected by COVID-19 infection. Therefore, early classification of it is a crucial task. Quantum machine learning and quantum neural network models can play an effective role in multiclass classification problems. Compared to standard deep and machine learning classifiers, the quantum variational classifier may lead to less memory usage, accuracy, and portability for respiratory disease detection. This article proposes a hybrid respiratory lung disease detection framework based on classical CNN and Quantum classifiers. It combines a classical deep feature extraction model with quantum classifiers. A new custom convolutional neural network (CCNN) deep learning model is proposed to perform feature extraction, and the Multi-Multi-Single (MMS) & Multi-Single-Multi-Single (MSMS) are proposed as quantum machine learning algorithms. These two quantum classifiers are built via a quantum variational circuit with encoding, entanglement, and measurement properties. The tests were carried out on the COVID-19 Radiography Dataset (CRD), which contains 15,153 chest X-ray images of COVID-19, Viral, and Normal. The experimental results revealed that the proposed model had the highest training and testing accuracy of 98.9% and 98.1%, on the CRD dataset, with a computation cost of 0.07 and 0.08 respectively. This hybrid model performs better than the other standard deep learning models. Additionally, we validated our MMS and MSMS quantum classifiers by deploying them on the IBM Q-QASM real-time quantum computer.

Deep learning-based cancer disease classification using Gene Expression Data

<p>Cancer disease caused a major death in worldwide. To prevent this, various cancer classification approaches are employed, which are mostly relied on clinical characteristics and histopathological characteristics. Deep learning-based classification models are very effective and accurate. As a result, this research established the Adam-based Deep Quantum Neural Network, which is the optimal deep learning-based cancer classification method. The information on gene expression is used to classify cancer. Using the Box-Cox transformation, which converts the data into a legible format, the data transformation procedure is carried out. Utilizing information gain, the features are chosen in order to choose the proper gene expression. Additionally, Deep QNN is used to classify cancer, and for better classification, which is trained via Adam optimization. The experimental result shows the developed model provide better classification result with respect to accuracy, true positive rate and true negative rate of 94.91%, 95.59% and 95.4%.</p>

Quantum Neural Networks in the Problem of Pattern Recognition

Three computational models are considered: classical, hybrid (NISQ) and quantum computational models, their pros, cons, possibilities of implementation in modern realities and the problem of image classification and its solution using neural networks in these computational models. Three computational experiments were carried out using the described image recognition approaches with visualization of the learning process and a comparison of the final metrics was carried out aimed at clarifying the prospects of the applied approach based on quantum computing.

Experimentally realizable continuous-variable quantum neural networks

Experimentally realizable continuous-variable quantum neural networks

Quantum machine learning based on continuous variable single-photon states: an elementary foundation for quantum neural networks

Quantum machine learning based on continuous variable single-photon states: an elementary foundation for quantum neural networks

A transient stability enhancement framework based on rapid fault-type identification for virtual synchronous generators

The virtual synchronous generator (VSG) control, widely used for voltage source inverters, is prone to the loss of synchronization (LOS) under large disturbances such as grid voltage sags. Despite the significant amount of research that has been conducted on the transient stability of VSGs, the existing conclusions on the influence of virtual inertia under different fault conditions remain controversial. In this paper, the conflicting effect of virtual inertia on VSG transient stability under different severities of grid voltage sag is comprehensively investigated. Faults, like grid voltage sags, are classified into two types in this paper: transient tolerable faults and transient intolerable faults. Moreover, a unified control framework for rapid fault-type identification and transient stability improvement is proposed. The Learning vector quantization (LVQ) neural networks are used to distinguish fault types within 5 ms. Based on the fault-type identification results, the proposed control strategies are applied to enhance the transient stability of VSG. Finally, simulations and control-hardware-in-loop experimental validations are carried out to demonstrate the effectiveness of the proposed framework.

Pulse-efficient quantum machine learning

Quantum machine learning algorithms based on parameterized quantum circuits are promising candidates for near-term quantum advantage. Although these algorithms are compatible with the current generation of quantum processors, device noise limits their performance, for example by inducing an exponential flattening of loss landscapes. Error suppression schemes such as dynamical decoupling and Pauli twirling alleviate this issue by reducing noise at the hardware level. A recent addition to this toolbox of techniques is pulse-efficient transpilation, which reduces circuit schedule duration by exploiting hardware-native cross-resonance interaction. In this work, we investigate the impact of pulse-efficient circuits on near-term algorithms for quantum machine learning. We report results for two standard experiments: binary classification on a synthetic dataset with quantum neural networks and handwritten digit recognition with quantum kernel estimation. In both cases, we find that pulse-efficient transpilation vastly reduces average circuit durations and, as a result, significantly improves classification accuracy. We conclude by applying pulse-efficient transpilation to the Hamiltonian Variational Ansatz and show that it delays the onset of noise-induced barren plateaus.

Problem-Dependent Power of Quantum Neural Networks on Multiclass Classification.

Quantum neural networks (QNNs) have become an important tool for understanding the physical world, but their advantages and limitations are not fully understood. Some QNNs with specific encoding methods can be efficiently simulated by classical surrogates, while others with quantum memory may perform better than classical classifiers. Here we systematically investigate the problem-dependent power of quantum neural classifiers (QCs) on multiclass classification tasks. Through the analysis of expected risk, a measure that weighs the training loss and the generalization error of a classifier jointly, we identify two key findings: first, the training loss dominates the power rather than the generalization ability; second, QCs undergo a U-shaped risk curve, in contrast to the double-descent risk curve of deep neural classifiers. We also reveal the intrinsic connection between optimal QCs and the Helstrom bound and the equiangular tight frame. Using these findings, we propose a method that exploits loss dynamics of QCs to estimate the optimal hyperparameter settings yielding the minimal risk. Numerical results demonstrate the effectiveness of our approach to explain the superiority of QCs over multilayer Perceptron on parity datasets and their limitations over convolutional neural networks on image datasets. Our work sheds light on the problem-dependent power of QNNs and offers a practical tool for evaluating their potential merit.

Geometric properties of quantum entanglement and machine learning

Objectives. Fast data analysis based on hidden patterns is one of the main issues for adaptive artificial intelligence systems development. This paper aims to propose and verify a method of such analysis based on the representation of data in the form of a quantum state, or, alternatively, in the form of a geometric object in a space allowing online machine learning.Methods. This paper uses Feynman formalism to represent quantum states and operations on them, the representation of quantum computing in the form of quantum circuits, geometric transformations, topological classification, as well as methods of classical and quantum machine learning. The Python programming language is used as a development tool. Optimization tools for machine learning are taken from the SciPy module. The datasets for analysis are taken from open sources. Data preprocessing was performed by the method of mapping features into numerical vectors, then the method of bringing the data to the desired dimension was applied. The data was then displayed in a quantum state. A proprietary quantum computing emulator is used (it is in the public domain).Results. The results of computational experiments revealed the ability of very simple quantum circuits to classify data without optimization. Comparative indicators of classification quality are obtained without the use of optimization, as well as with its use. Experiments were carried out with different datasets and for different values of the dimension of feature spaces. The efficiency of the models and methods of machine learning proposed in the work, as well as methods of combining them into network structures, is practically confirmed.Conclusions. The proposed method of machine learning and the model of quantum neural networks can be used to create adaptive artificial intelligence systems as part of an online learning module. Free online optimization learning process allows it to be applied in data streaming, that is, adapting to changes in the environment. The developed software does not require quantum computers and can be used in the development of artificial intelligence systems in Python as imported modules.

A fractional-order quantum neural network: dynamics, finite-time synchronization

A model of fractional-order quantum cellular neural network (FoQCNN) by using fractional-order quantum-dot cellular automata (QCA) is constructed and its dynamics are analyzed. Then, a robust finite-time synchronization scheme using terminal sliding mode control (SMC) technique is proposed. And then, taking the perturbed FoQCNN model with uncertainties and external disturbances as an example, the results are simulated which present the proposed scheme is effective. It has robust synchronization performance and good anti-interference ability, which provides a theoretical basis for the application of neural network in security.

Recent Advances for Quantum Neural Networks in Generative Learning.

Quantum computers are next-generation devices that hold promise to perform calculations beyond the reach of classical computers. A leading method towards achieving this goal is through quantum machine learning, especially quantum generative learning. Due to the intrinsic probabilistic nature of quantum mechanics, it is reasonable to postulate that quantum generative learning models (QGLMs) may surpass their classical counterparts. As such, QGLMs are receiving growing attention from the quantum physics and computer science communities, where various QGLMs that can be efficiently implemented on near-term quantum machines with potential computational advantages are proposed. In this paper, we review the current progress of QGLMs from the perspective of machine learning. Particularly, we interpret these QGLMs, covering quantum circuit Born machines, quantum generative adversarial networks, quantum Boltzmann machines, and quantum variational autoencoders, as the quantum extension of classical generative learning models. In this context, we explore their intrinsic relations and their fundamental differences. We further summarize the potential applications of QGLMs in both conventional machine learning tasks and quantum physics. Last, we discuss the challenges and further research directions for QGLMs.

К ВОПРОСУ КВАНТОВОЙ ТЕОРИИ ФИЗИКИ МОЗГА И ТЕОРИИ КВАНТОВО-ПОДОБНОГО СОЗНАНИЯ (ОБЗОР ЗАРУБЕЖНОЙ ЛИТЕРАТУРЫ)

The article provides a review of foreign scientific literature on existing approaches in scientists' ideas about the quantum theory of brain physics and the theory of quantum-like consciousness. It is shown that in recent years, physics and psychology have made important scientific advances; the emergence of new tools has led to the development of neuroscience, which makes it possible to study the state of the brain when known mental functions are preserved or impaired. The use of mathematical methods has made it possible to develop computational models that describe the fundamental functions of the brain and implement them in technological applications. The most pressing issues of modern research in the field of mental processes of the brain, such as perception, cognition, emotions, consciousness and learning, researchers call the problems of the universal properties of human behavior in various conditions and cultures; the potential of computational psychiatry to predict, understand and treat mental disorders; neural mechanisms of influence on mental content at higher levels of organization of the hierarchical brain; mechanisms of the emergence of abstract ideas from sensorimotor accidents, etc. Currently, in understanding the cognitive processes of the brain, perception, understanding, cognition, mental lexicon, understanding the generation and nature of emotions, memory, learning, etc. A number of researchers use the theory of quantum or quantum-like consciousness. Research in this area is both empirical and theoretical in nature, with the postulates of the theory being considered mainly related to quantum physics due to the lack of other areas in which quantum-like properties are supposedly manifested. It is believed that despite modern advances in artificial intelligence models, neural networks cannot reach the level of the human brain, partly due to differences in the way information is encoded and processed. In order to develop the potential of brain-based artificial intelligence, quantum neural networks are proposed that are similar to the mechanisms and phenomena in the brain. Interactions between biologically important molecules in the brain may underlie the quantum metalanguage that governs the behavior of humans and animals.

ИЗЫСКАНИЕ ОРИГИНАЛЬНОЙ МОДЕЛИ ДЛЯ ИЗУЧЕНИЯ ВЛИЯНИЯ ФАРМАКОЛОГИЧЕСКИ АКТИВНЫХ ВЕЩЕСТВ НА СКОРОСТЬ ПОТРЕБЛЕНИЯ КИСЛОРОДА ТКАНЬЮ ПРЕДСТАТЕЛЬНОЙ ЖЕЛЕЗЫ

The article provides a review of foreign scientific literature on existing approaches in scientists' ideas about the quantum theory of brain physics and the theory of quantum-like consciousness. It is shown that in recent years, physics and psychology have made important scientific advances; the emergence of new tools has led to the development of neuroscience, which makes it possible to study the state of the brain when known mental functions are preserved or impaired. The use of mathematical methods has made it possible to develop computational models that describe the fundamental functions of the brain and implement them in technological applications. The most pressing issues of modern research in the field of mental processes of the brain, such as perception, cognition, emotions, consciousness and learning, researchers call the problems of the universal properties of human behavior in various conditions and cultures; the potential of computational psychiatry to predict, understand and treat mental disorders; neural mechanisms of influence on mental content at higher levels of organization of the hierarchical brain; mechanisms of the emergence of abstract ideas from sensorimotor accidents, etc. Currently, in understanding the cognitive processes of the brain, perception, understanding, cognition, mental lexicon, understanding the generation and nature of emotions, memory, learning, etc. A number of researchers use the theory of quantum or quantum-like consciousness. Research in this area is both empirical and theoretical in nature, with the postulates of the theory being considered mainly related to quantum physics due to the lack of other areas in which quantum-like properties are supposedly manifested. It is believed that despite modern advances in artificial intelligence models, neural networks cannot reach the level of the human brain, partly due to differences in the way information is encoded and processed. In order to develop the potential of brain-based artificial intelligence, quantum neural networks are proposed that are similar to the mechanisms and phenomena in the brain. Interactions between biologically important molecules in the brain may underlie the quantum metalanguage that governs the behavior of humans and animals.

EQNAS: Evolutionary Quantum Neural Architecture Search for Image Classification

Quantum neural network (QNN) is a neural network model based on the principles of quantum mechanics. The advantages of faster computing speed, higher memory capacity, smaller network size and elimination of catastrophic amnesia make it a new idea to solve the problem of training massive data that is difficult for classical neural networks. However, the quantum circuit of QNN are artificially designed with high circuit complexity and low precision in classification tasks. In this paper, a neural architecture search method EQNAS is proposed to improve QNN. First, initializing the quantum population after image quantum encoding. The next step is observing the quantum population and evaluating the fitness. The last is updating the quantum population. Quantum rotation gate update, quantum circuit construction and entirety interference crossover are specific operations. The last two steps need to be carried out iteratively until a satisfactory fitness is achieved. After a lot of experiments on the searched quantum neural networks, the feasibility and effectiveness of the algorithm proposed in this paper are proved, and the searched QNN is obviously better than the original algorithm. The classification accuracy on the mnist dataset and the warship dataset not only increased by 5.31% and 4.52%, respectively, but also reduced the parameters by 21.88% and 31.25% respectively. Code will be available at https://gitee.com/Pcyslist/models/tree/master/research/cv/EQNAS, and https://github.com/Pcyslist/EQNAS.

Resource saving via ensemble techniques for quantum neural networks

Quantum neural networks hold significant promise for numerous applications, particularly as they can be executed on the current generation of quantum hardware. However, due to limited qubits or hardware noise, conducting large-scale experiments often requires significant resources. Moreover, the output of the model is susceptible to corruption by quantum hardware noise. To address this issue, we propose the use of ensemble techniques, which involve constructing a single machine learning model based on multiple instances of quantum neural networks. In particular, we implement bagging and AdaBoost techniques, with different data loading configurations, and evaluate their performance on both synthetic and real-world classification and regression tasks. To assess the potential performance improvement under different environments, we conducted experiments on both simulated, noiseless software and IBM superconducting-based QPUs, suggesting these techniques can mitigate the quantum hardware noise. Additionally, we quantify the amount of resources saved using these ensemble techniques. Our findings indicate that these methods enable the construction of large, powerful models even on relatively small quantum devices.

Deep learning models for price forecasting of financial time series: A review of recent advancements: 2020–2022

AbstractAccurately predicting the prices of financial time series is essential and challenging for the financial sector. Owing to recent advancements in deep learning techniques, deep learning models are gradually replacing traditional statistical and machine learning models as the first choice for price forecasting tasks. This shift in model selection has led to a notable rise in research related to applying deep learning models to price forecasting, resulting in a rapid accumulation of new knowledge. Therefore, we conducted a literature review of relevant studies over the past 3 years with a view to aiding researchers and practitioners in the field. This review delves deeply into deep learning‐based forecasting models, presenting information on model architectures, practical applications, and their respective advantages and disadvantages. In particular, detailed information is provided on advanced models for price forecasting, such as Transformers, generative adversarial networks (GANs), graph neural networks (GNNs), and deep quantum neural networks (DQNNs). The present contribution also includes potential directions for future research, such as examining the effectiveness of deep learning models with complex structures for price forecasting, extending from point prediction to interval prediction using deep learning models, scrutinizing the reliability and validity of decomposition ensembles, and exploring the influence of data volume on model performance.This article is categorized under: Technologies > Prediction Technologies > Artificial Intelligence

BISDU: A Bit-Serial Dot-Product Unit for Microcontrollers

Low-precision quantized neural networks (QNNs) reduce the required memory space, bandwidth, and computational power, and hence are suitable for deployment in applications such as IoT edge devices. Mixed-precision QNNs, where weights commonly have lower precision than activations or different precision is used for different layers, can limit the accuracy loss caused by low-bit quantization, while still benefiting from reduced memory footprint and faster execution. Previous multiple-precision functional units supporting 8-bit, 4-bit, and 2-bit SIMD instructions have limitations, such as large area overhead, under-utilization of multipliers, and wasted memory space for low and mixed bit-width operations. This article introduces BISDU, a bit-serial dot-product unit to support and accelerate execution of mixed-precision low-bit QNNs on resource-constrained microcontrollers. BISDU is a multiplier-less dot-product unit, with frugal hardware requirements (a population count unit and 2:1 multiplexers). The proposed bit-serial dot-product unit leverages the conventional logical operations of a microcontroller to perform multiplications, which enables efficient software implementations of binary ( Xnor ), ternary ( Xor ), and mixed-precision [W×A] ( And ) dot-product operations. The experimental results show that BISDU achieves competitive performance compared to two state-of-the-art units, XpulpNN and Dustin, when executing low-bit-width CNNs. We demonstrate the advantage that bit-serial execution provides by enabling trading accuracy against weight footprint and execution time. BISDU increases the area of the ALU by 68% and the ALU power consumption by 42% compared to a baseline 32-bit RISC-V (RV32IC) microcontroller core. In comparison, XpulpNN and Dustin increase the area by 6.9× and 11.1× and the power consumption by 3.8× and 5.97×, respectively. The bit-serial state-of-the-art, based on a conventional popcount instruction, increases the area by 42% and power by 32%, with BISDU providing a 37% speedup over it.

Unsupervised machine learning and multi-seismic attributes for fault and fracture network interpretation in the Kerry Field, Taranaki Basin, New Zealand

Unsupervised machine learning using an unsupervised vector quantization neural network (UVQ-NN) integrated with meta-geometrical attributes as a novel computation process as opposed to traditional methodologies is currently used effectively in the 3D seismic structural interpretation for high-resolution detection of fault patterns, fracture network zones, and small-scale faults (SSFs). This technology has a crucial role in locating prospective well sites and building a 3D structural model while saving time and cost. The innovation of the current workflow involves combining geostatistical and structural filtering, optimal geometrical seismic attributes, UVQ-NN for automatic major faults, fracture network zones, and SSFs volumes extraction due to the unavailability of well logs and cores. To sharpen the fault edges and discontinuities, a steered volume was first extracted. Structural filters were then applied to the 3D volume, first with a dip-steered median filter (DSMF), followed by a dip-steered diffusion filter (DSDF), and finally, both DSMF and DSDF were combined to generate the fault enhancement filter (FEF). After that, optimal geometrical attributes were computed and extracted, such as similarity, FEF on similarity, maximum curvature, polar dip, fracture density, and thinned fault likelihood (TFL) attributes. Finally, selected attributes were inserted as the input layer to the UVQ-NN to generate segmentation and matching volumes. On the other hand, the TFL was used with the voxel connectivity filter (VCF) for 3D automatic fault patches extraction. The results from the UVQ-NN and VCF identified the locations, orientations, and extensions of the main faults, SSFs, and fracture networks. The implemented approach is innovative and can be employed in the future for the identification, extraction, and classification of geological faults and fracture networks in any region of the world.

Toward improving the performance of learning by joining feature selection and ensemble classification techniques: an application for cancer diagnosis.

Breast cancer is known as the most common type of cancer in women, and this has raised the importance of its diagnosis in medical science as one of the most important issues. In addition to reducing costs, the diagnosis of benign or malignant breast cancer is very important in determining the treatment method. The purpose of this paper is to present a model based on data mining techniques including feature selection and ensemble classification that can accurately predict breast cancer patients in the early stages. The proposed breast cancer detection model is developed by joining Adaptive Differential Evolution (ADE) algorithm for feature selection and Learning Vector Quantization (LVQ) neural network for classification. Our proposed model as ADE-LVQ has the ability to automatically and quickly diagnose breast cancer patients into two classes, benign and malignant. As a new evolutionary approach, ADE performs optimal configuration for LVQ neural network in addition to selecting effective features from breast cancer data. Meanwhile, we configure an ensemble classification technique based on LVQ, which significantly improves the prediction performance. ADE-LVQ has been analyzed from different perspectives on different datasets from Wisconsin breast cancer database. We apply different approaches to handle missing values and improve data quality on this database. The results of the simulations showed that the ADE-LVQ model is more successful than the equivalent and state-of-the-art models in diagnosing breast cancer patients. Also, ADE-LVQ provides better performance with less complexity, considering feature selection and ensemble learning. In particular, ADE-LVQ improves accuracy (up to 3.4%) and runtime (up to 2.3%) on average compared to the existing best method. Combined methods based on data mining techniques for breast cancer diagnosis can help doctors in making better decisions for disease treatment.