Article
Quantum Science & Technology
Carsten Blank, Adenilton J. da Silva, Lucas P. de Albuquerque, Francesco Petruccione, Daniel K. Park
Summary: Quantum computing offers exciting opportunities for kernel-based machine learning methods, allowing efficient construction of classifier models through quantum interference effects. To make these methods practical, it is important to minimize circuit size and handle imbalanced data sets.
QUANTUM SCIENCE AND TECHNOLOGY
(2022)
Article
Automation & Control Systems
Mingxuan Liang, Kai Zhou
Summary: A probabilistic fault diagnosis framework using a Gaussian process classifier was developed to account for uncertainties in prediction, showcasing unique capability and outperforming other machine learning models in accuracy and robustness in systematic case studies. Sensor fusion with spatial vibration measurements further enhances fault diagnosis performance. Leveraging the probabilistic feature facilitates future research on extended fault diagnosis using limited fault labels.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Quantum Science & Technology
Jahan Claes, J. Eli Bourassa, Shruti Puri
Summary: Fault-tolerant cluster states are crucial for scalable measurement-based quantum computation. Although high-threshold stabilizer codes have been introduced for biased noise in circuit-based quantum computation, extending these codes to generate high-threshold cluster states for biased noise has been challenging. In this work, we introduce a generalization of cluster states that enables foliating stabilizer codes while preserving the noise bias. Our XZZX cluster state, constructed using this approach, shows a threshold more than double the usual cluster state when dephasing errors are more likely than bit-flip errors by a factor of order similar to 100 or more.
NPJ QUANTUM INFORMATION
(2023)
Article
Physics, Multidisciplinary
Takafumi Ono, Wojciech Roga, Kentaro Wakui, Mikio Fujiwara, Shigehito Miki, Hirotaka Terai, Masahiro Takeoka
Summary: In this study, a new quantum classifier for bosonic systems is proposed and successfully demonstrated using a silicon-based photonic integrated circuit. By implementing a programmable optical circuit combined with an interferometer, a high success probability in classification is achieved in the proof of principle experiment.
PHYSICAL REVIEW LETTERS
(2023)
Article
Physics, Multidisciplinary
Takafumi Ono, Wojciech Roga, Kentaro Wakui, Mikio Fujiwara, Shigehito Miki, Hirotaka Terai, Masahiro Takeoka
Summary: In this study, we propose a new quantum classifier for bosonic systems using the data reuploading technique and demonstrate it with a silicon-based photonic integrated circuit. By implementing a programmable optical circuit combined with an interferometer, we achieve a classification success probability of 94.8% in the proof of principle experiment with uncorrelated two photons. This method has the potential for further development in optical quantum classifiers, including extensions to quantum entangled and multiphoton states.
PHYSICAL REVIEW LETTERS
(2023)
Article
Biochemistry & Molecular Biology
Jonathan Kim, Stefan Bekiranov
Summary: Quantum metric learning is a method that can classify data into categories by learning through quantum embedding, achieving classification in high-dimensional feature data. The research found that by reducing data dimensionality and limiting the number of model parameters, quantum metric learning can accurately classify test data.
Article
Physics, Applied
Rui Yang, Samuel Bosch, Bobak Kiani, Seth Lloyd, Adrian Lupascu
Summary: Quantum machine learning has the potential to provide powerful algorithms for artificial intelligence, and various quantum-classical hybrid algorithms have been proposed. This study proposes a quantum variational embedding classifier based on an analog quantum computer, where classical data are transformed into the parameters of the time-varying Hamiltonian. The nonlinear dependence of the final quantum state on the control parameters of the Hamiltonian provides the nonlinearity needed for a nonlinear classification problem. The algorithm shows promise for using current quantum annealers for practical machine learning problems and exploring quantum advantage in quantum machine learning.
PHYSICAL REVIEW APPLIED
(2023)
Article
Quantum Science & Technology
Jie Zhou, Dongfen Li, Yuqiao Tan, Xiaolong Yang, Yundan Zheng, Xiaofang Liu
Summary: This paper proposes a new variational quantum multi-class classifier that achieves high accuracy in classification by using quantum bits to represent labels and optimizing circuit parameters.
QUANTUM INFORMATION PROCESSING
(2023)
Article
Physics, Multidisciplinary
Yuan Li, Yinkuo Meng, Yiyuan Luo
Summary: A quantum version of classifier based on physical graph state is proposed, which is built through an entangling-subgraphs process. Leveraging the efficiency of graph state, the quantum algorithm is more suitable for processing big data compared to existing classical methods.
INTERNATIONAL JOURNAL OF THEORETICAL PHYSICS
(2021)
Article
Multidisciplinary Sciences
Giuseppe Sergioli, Carmelo Militello, Leonardo Rundo, Luigi Minafra, Filippo Torrisi, Giorgio Russo, Keng Loon Chow, Roberto Giuntini
Summary: Recent advances in Quantum Machine Learning (QML) have shown benefits in reducing time complexity. Another approach, Quantum-inspired Machine Learning (QiML), uses the expressive power of quantum language to increase accuracy. The proposed experiment used a quantum-inspired binary classifier to improve clonogenic assay evaluation in biomedical imaging, showing homogeneity as a relevant feature in detecting challenging cell colonies. The novel quantum-inspired classifier outperformed conventional machine learning classifiers in this context.
SCIENTIFIC REPORTS
(2021)
Article
Quantum Science & Technology
David Bauch, Dustin Siebert, Klaus D. Joens, Jens Foerstner, Stefan Schumacher
Summary: This study focuses on generating photon pairs with high degrees of polarization entanglement and high indistinguishability. It achieves this by selectively reducing the biexciton lifetime with an optical resonator. Through the optimization of photonic structures and microscopic simulations of quantum-dot cavity excitation dynamics, it determines the optimal range of Purcell enhancement for maximizing indistinguishability and entanglement.
ADVANCED QUANTUM TECHNOLOGIES
(2023)
Article
Computer Science, Information Systems
Dong Wei, Xiaobo Shen, Quansen Sun, Xizhan Gao, Zhenwen Ren
Summary: This study presents two metric learning algorithms based on Grassmann manifold for image set classification and exploring intrinsic geometry distance. The proposed algorithms perform favorably against the state-of-the-art methods in extensive experiments.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Physics, Multidisciplinary
Yuan Li, Duan Huang
Summary: In this paper, a novel graphical encoding method is applied to establish the mapping between the feature space of sample data and the two-level nested graph state in quantum computing, enabling the realization of a binary quantum classifier for large-scale test states. The proposed boosting algorithm adjusts weights to address classification errors caused by noise, resulting in a strong classifier with significantly improved accuracy. Experimental investigation demonstrates the superiority of the proposed algorithm in certain aspects. This work enriches the theoretical foundation of quantum graph theory and quantum machine learning, offering potential for assisting the classification of massive-data networks through entangled subgraphs.
Article
Quantum Science & Technology
Jonathan F. San Miguel, Dominic J. Williamson, Benjamin J. Brown
Summary: In this study, a cellular automaton decoder for the XYZ color code is proposed, where the bases of the physical qubits are locally rotated. The decoder demonstrates the key properties of a two-dimensional fractal code if the noise is biased towards dephasing. The results suggest the design of tailored cellular automaton decoders to reduce the bandwidth demands of global decoding for realistic noise models.
Article
Physics, Multidisciplinary
Weikang Li, Sirui Lu, Dong-Ling Deng
Summary: Private distributed learning explores collaborative training of deep networks with private data using quantum protocols, showing potential for handling computationally expensive tasks with privacy guarantees.
SCIENCE CHINA-PHYSICS MECHANICS & ASTRONOMY
(2021)
Article
Chemistry, Analytical
Anton V. Bourdine, Vladimir V. Demidov, Artem A. Kuznetsov, Alexander A. Vasilets, Egishe V. Ter-Nersesyants, Alexander V. Khokhlov, Alexandra S. Matrosova, Grigori A. Pchelkin, Michael V. Dashkov, Elena S. Zaitseva, Azat R. Gizatulin, Ivan K. Meshkov, Airat Zh. Sakhabutdinov, Eugeniy V. Dmitriev, Oleg G. Morozov, Vladimir A. Burdin, Konstantin V. Dukelskii, Yaseera Ismail, Francesco Petruccione, Ghanshyam Singh, Manish Tiwari, Juan Yin
Summary: This work presents the design and fabrication of a silica few-mode optical fiber with induced twisting and improved refractive index profile. The fiber supports 4 guided modes over the C-band and has been tested for its mode properties after fiber Bragg grating writing.
Article
Physics, Multidisciplinary
Jihye Kim, Byungdu Oh, Yonuk Chong, Euyheon Hwang, Daniel K. Park
Summary: In this work, a deep learning-based protocol is presented for reducing readout errors on quantum hardware. By training a neural network to correct non-linear noise, the limitations of existing linear inversion methods are overcome.
NEW JOURNAL OF PHYSICS
(2022)
Article
Quantum Science & Technology
Carsten Blank, Adenilton J. da Silva, Lucas P. de Albuquerque, Francesco Petruccione, Daniel K. Park
Summary: Quantum computing offers exciting opportunities for kernel-based machine learning methods, allowing efficient construction of classifier models through quantum interference effects. To make these methods practical, it is important to minimize circuit size and handle imbalanced data sets.
QUANTUM SCIENCE AND TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Roberto Giuntini, Federico Holik, Daniel K. Park, Hector Freytes, Carsten Blank, Giuseppe Sergioli
Summary: Quantum machine learning is a groundbreaking discipline that exploits the peculiarities of quantum computation for machine learning tasks. Quantum-inspired machine learning has demonstrated relevant benefits for machine learning problems without employing quantum computers. This study introduces a quantum-inspired classifier for multi-class classification that outperforms standard binary classifiers in terms of accuracy and time complexity.
APPLIED SOFT COMPUTING
(2023)
Article
Optics
Shweta Mittal, Ankur Saharia, Yaseera Ismail, Francesco Petruccione, Anton V. Bourdine, Oleg G. Morozov, Vladimir V. Demidov, Juan Yin, Ghanshyam Singh, Manish Tiwari
Summary: This work presents the design and simulation of an all-optical sensor based on surface plasmon resonance effect on a spiral shaped photonic crystal fiber structure for detection of different cancer cells. The sensor showed high sensitivity and resolution for detecting breast cancer cells, with potential applications for detecting other types of cancer such as cervical cancer, skin cancer, blood cancer, and adrenal gland cancer.
Article
Materials Science, Multidisciplinary
Anton V. Bourdine, Vladimir V. Demidov, Konstantin V. Dukelskii, Alexander V. Khokhlov, Egishe V. Ter-Nersesyants, Sergei V. Bureev, Alexandra S. Matrosova, Grigori A. Pchelkin, Artem A. Kuznetsov, Oleg G. Morozov, Ilnur I. Nureev, Airat Zh. Sakhabutdinov, Timur A. Agliullin, Michael V. Dashkov, Alexander S. Evtushenko, Elena S. Zaitseva, Alexander A. Vasilets, Azat R. Gizatulin, Ivan K. Meshkov, Yaseera Ismail, Francesco Petruccione, Ghanshyam Singh, Manish Tiwari, Juan Yin
Summary: This article presents a fabricated silica few-mode microstructured optical fiber (MOF) with a special six GeO2-doped core geometry. The fiber has an outer diameter of 125 mu m and improved induced twisting up to 500 revolutions per 1 m. The article discusses the technological aspects and issues of manufacturing twisted MOFs with complicated structures and geometry.
Article
Quantum Science & Technology
Matt Lourens, Ilya Sinayskiy, Daniel K. Park, Carsten Blank, Francesco Petruccione
Summary: Quantum circuit algorithms, like neural and tensor networks, require hierarchical, modular and repeating architectural design choices. Neural Architecture Search (NAS) automates neural network design and achieves state-of-the-art performance. We propose a framework for representing quantum circuit architectures using NAS techniques, enabling search space design and architecture search. We demonstrate the importance of circuit architecture in quantum machine learning by generating Quantum Convolutional Neural Networks (QCNNs) and evaluating them on a music genre classification dataset. We also employ a genetic algorithm for Quantum Phase Recognition (QPR) as an example of architecture search, and provide an open-source Python package for dynamic circuit creation and NAS circuit search space design.
NPJ QUANTUM INFORMATION
(2023)
Article
Quantum Science & Technology
Israel F. Araujo, Daniel K. Park, Teresa B. Ludermir, Wilson R. Oliveira, Francesco Petruccione, Adenilton J. da Silva
Summary: The theory of quantum algorithms promises the benefits of using the laws of quantum mechanics to solve computational problems. However, a prerequisite for applying these algorithms is loading classical data onto a quantum state. Existing methods either require linear growth in quantum circuit depth or width, nullifying the advantage of representing exponentially many classical data in a quantum state. This paper presents a configurable bidirectional procedure that balances the trade-off between quantum circuit width and depth, allowing for sublinear growth when encoding N-dimensional classical data.
QUANTUM INFORMATION PROCESSING
(2023)
Article
Optics
Adenilton J. da Silva, Daniel K. Park
Summary: The paper presents a systematic procedure for decomposing multiqubit controlled unitary gates into controlled -NOT and single-qubit gates to minimize quantum circuit depth. The algorithm does not require ancillary qubits and achieves a quadratic reduction in circuit depth compared to known methods.
Article
Physics, Multidisciplinary
Kimara Naicker, Ilya Sinayskiy, Francesco Petruccione
Summary: The hierarchical equations of motion (HEOM) are used to simulate the dynamics of an open quantum system, and a classical machine learning (ML) approach is employed to solve the computational problem. The ML models, including convolutional neural networks, are capable of accurately predicting Hamiltonian parameters with a 99.28% accuracy rate.
PHYSICAL REVIEW RESEARCH
(2022)
Article
Optics
Vinayak Jagadish, R. Srikanth, Francesco Petruccione
Summary: This article studies the convex combinations of (d+1)-generalized Pauli dynamical maps in a Hilbert space of dimension d. It is found that for certain choices of the decoherence function, the maps are noninvertible and this noninvertibility remains under convex combinations. We evaluate the fraction of invertible maps obtained upon mixing for dynamical maps characterized by a specific decoherence function, and observe that this fraction increases superexponentially with dimension d.
Article
Optics
Vinayak Jagadish, R. Srikanth, Francesco Petruccione
Summary: This study investigates the conditions under which a semigroup is obtained by convex combinations of channels, specifically focusing on the set of Pauli and generalized Pauli channels. The findings show that merely mixing semigroups cannot result in a semigroup. Contrary to intuition, it is discovered that for a convex combination to yield a semigroup, the majority of input channels must be noninvertible.
Proceedings Paper
Computer Science, Artificial Intelligence
Shivani Mahashakti Pillay, Ilya Sinayskiy, Edgar Jembere, Francesco Petruccione
Summary: This study demonstrates the principle of quantum-kernel-based classifiers applied to non-linearly separable datasets. By applying different post-processing strategies to the kernel matrices, the accuracy of the classifiers can be improved. Quantum-kernel-based classifiers show high effectiveness in the Noisy Intermediate Scale Quantum (NISQ) computing era.
ARTIFICIAL INTELLIGENCE RESEARCH, SACAIR 2021
(2022)
Article
Computer Science, Artificial Intelligence
Tak Hur, Leeseok Kim, Daniel K. Park
Summary: This work benchmarks the performance of parameterized quantum convolutional neural networks (QCNNs) for classical data classification. The QCNN models, inspired by CNN, utilize two-qubit interactions and achieve excellent classification accuracy on MNIST and Fashion MNIST datasets. Compared to CNN models under similar training conditions, QCNN models perform better and are suitable for NISQ quantum devices.
QUANTUM MACHINE INTELLIGENCE
(2022)