摘要 :
The aim of this work is to give an introduction for a non-practical reader to the growing field of quantum machine learning, which is a recent discipline that combines the research areas of machine learning and quantum computing. ...
展开
The aim of this work is to give an introduction for a non-practical reader to the growing field of quantum machine learning, which is a recent discipline that combines the research areas of machine learning and quantum computing. This work presents the most notable scientific literature about quantum machine learning, starting from the basics of quantum logic to some specific elements and algorithms of quantum computing (such as QRAM, Grover and HHL), in order to allow a better understanding of latest quantum machine learning techniques. The main aspects of quantum machine learning are then covered, with detailed descriptions of some notable algorithms, such as quantum natural gradient and quantum support vector machines, up to the most recent quantum deep learning techniques, such as quantum neural networks.
收起
Machine learning is a branch of artificial intelligence, and it has been widely used in many science and engineering areas, such as data mining, natural language processing, computer vision, biological ana
Machine learning is a branch of artificial intelligence, and it has been widely used in many science and engineering areas, such as data mining, natural language processing, computer vision, biological analysis, and so on. Quantum computer is considered as one of the most promising technologies of human beings in the near future. With the development of machine learning and quantum computing, researchers consider to combine these two aspects to gain more benefits. As a result, a novel interdisciplinary subject has emerged—quantum machine learning. This article reviews the state‐of‐the‐art research of algorithms of quantum machine learning and shows a path of the research from the basic quantum information to quantum machine learning algorithms from the perspective of people in the field of computer science.
摘要 :
In recent years, Quantum Computing witnessed massive improvements in terms of available resources and algorithms development. The ability to harness quantum phenomena to solve computational problems is a long-standing dream that h...
展开
In recent years, Quantum Computing witnessed massive improvements in terms of available resources and algorithms development. The ability to harness quantum phenomena to solve computational problems is a long-standing dream that has drawn the scientific community's interest since the late '80s. In such a context, we propose our contribution. First, we introduce basic concepts related to quantum computations, and then we explain the core functionalities of technologies that implement the Gate Model and Adiabatic Quantum Computing paradigms. Finally, we gather, compare, and analyze the current state-of-the-art concerning Quantum Perceptrons and Quantum Neural Networks implementations.
收起
摘要 :
Abstract Quantum machine learning (QML) has emerged as a promising domain offering significant computational advantages over classical counterparts. In recent times, researchers have directed their attention towards this field. Th...
展开
Abstract Quantum machine learning (QML) has emerged as a promising domain offering significant computational advantages over classical counterparts. In recent times, researchers have directed their attention towards this field. The objective of this paper is to provide a thorough overview of the advancements in quantum machine learning, encompassing the state-of-the-art approaches. The machine learning field is itself quite diverse. Diversity of QML is broadened due to the respective roles the quantum information processing and machine learning play in it. The study focuses on analysing the predictive efficacy of deep learning models on time series data. After experimental evaluation, we have chosen deep learning models that have better performance on time series data. The paper illustrates how different quantum techniques such as quantum encoding, optimization, etc., are used in quantum-enhanced models and provides a comprehensive review and an experimental analysis of three state-of-the-art quantum-enhanced models. Mental health is a serious global public health concern that has permeated modern civilization. So, seven time series data related to mental health conditions were collected from SWELL-KW, Wesad and psykose. Based on the experimental findings from the current dataset, it is evident that the quantum LSTM model exhibits superior predictive performance compared to other state-of-the-art approaches.
收起
摘要 :
Cloning an unknown state is an important task in the field of quantum computation as it is one of the basic operations required in any experiment. The no-cloning theorem states that it is impossible to create an identical copy of ...
展开
Cloning an unknown state is an important task in the field of quantum computation as it is one of the basic operations required in any experiment. The no-cloning theorem states that it is impossible to create an identical copy of an arbitrary unknown quantum state. Hence, techniques are developed to clone unknown states to high fidelities, rather than to exact copies. The usual method of cloning is quantum tomography, which measures a set of observables to reconstruct the unknown state. This method proves to be very expensive when the number of copies of the unknown state is limited. Here, we try to clone an unknown state in IBM's QASM simulator using a quantum reinforcement learning protocol (Albarran-Arriagada et al. in Phys Rev A 98:042315, 2018), where the "right" amount of punishment/reward function and boundary conditions can give much better fidelity than what tomography can offer in limited copies of the state. Using this method, we can attain above 90% fidelity in under 50 copies. This method proves to be very useful for reconstructing quantum states when only limited copies of the state are available.
收起
摘要 :
Training quantum neural networks (QNNs) using gradient-based or gradient-free classical optimization approaches is severely impacted by the presence of barren plateaus in the cost landscapes. In this paper, we devise a framework f...
展开
Training quantum neural networks (QNNs) using gradient-based or gradient-free classical optimization approaches is severely impacted by the presence of barren plateaus in the cost landscapes. In this paper, we devise a framework for leveraging quantum optimization algorithms to find optimal parameters of QNNs for certain tasks. To cast the optimization problem of training QNN into the context of quantum optimization, the parameters in QNN are quantized - moved from being classical to being stored in quantum registers which are in addition to those upon which the QNN is performing its computation. We then coherently encode the cost function of QNNs onto relative phases of a superposition state in the Hilbert space of the QNN parameters. The parameters are tuned with an iterative quantum optimization structure using adaptively selected Hamiltonians. The quantum mechanism of this framework exploits hidden structure in the QNN optimization problem and hence is expected to provide beyond-Grover speed up, mitigating the barren plateau issue.
收起
摘要 :
Traditional machine learning shares several benefits with quantum information processing field. The study of machine learning with quantum mechanics is called quantum machine learning. Data clustering is an important tool for mach...
展开
Traditional machine learning shares several benefits with quantum information processing field. The study of machine learning with quantum mechanics is called quantum machine learning. Data clustering is an important tool for machine learning where quantum computing plays a vital role in its inherent speed up capability. In this paper, a hybrid quantum algorithm for data clustering (quantum walk-based hybrid clustering (QWBHC)) is introduced where one-dimensional discrete time quantum walks (DTQW) play the central role to update the positions of data points according to their probability distributions. A quantum oracle is also designed and it is mainly implemented on a finite d-regular bipartite graph where data points are initially distributed as a predefined set of clusters. An overview of a quantum walk (QW) based clustering algorithm on 1D lattice structure is also introduced and described in this paper. In order to search the nearest neighbors, a unitary and reversible DTQW gives a quadratic speed up over the traditional classical random walk. This paper also demonstrates the comparisons of our proposed hybrid quantum clustering algorithm with some state-of-the-art clustering algorithms in terms of clustering accuracy and time complexity analysis. The proposed quantum oracle needs O(root n) queries to mark the nearest data points among clusters and modify the existing clusters. Finally, the proposed QWBHC algorithm achieves O(n) + O(root n) performance.
收起
摘要 :
Adversarial transfer learning is a machine learning method that employs an adversarial training process to learn the datasets of different domains. Recently, this method has attracted attention because it can efficiently decouple ...
展开
Adversarial transfer learning is a machine learning method that employs an adversarial training process to learn the datasets of different domains. Recently, this method has attracted attention because it can efficiently decouple the requirements of tasks from insufficient target data. In this study, we introduce the notion of quantum adversarial transfer learning, where data are completely encoded by quantum states. A measurement-based judgment of the data label and a quantum subroutine to compute the gradients are discussed in detail. We also prove that our proposal has an exponential advantage over its classical counterparts in terms of computing resources such as the gate number of the circuits and the size of the storage required for the generated data. Finally, numerical experiments demonstrate that our model can be successfully trained, achieving high accuracy on certain datasets.
收起
摘要 :
Anomaly detection is used for identifying data that deviate from "normal" data patterns. Its usage on classical data finds diverse applications in many important areas such as finance, fraud detection, medical diagnoses, data clea...
展开
Anomaly detection is used for identifying data that deviate from "normal" data patterns. Its usage on classical data finds diverse applications in many important areas such as finance, fraud detection, medical diagnoses, data cleaning, and surveillance. With the advent of quantum technologies, anomaly detection of quantum data, in the form of quantum states, may become an important component of quantum applications. Machine-learning algorithms are playing pivotal roles in anomaly detection using classical data. Two widely used algorithms are the kernel principal component analysis and the one-class support vector machine. We find corresponding quantum algorithms to detect anomalies in quantum states. We show that these two quantum algorithms can be performed using resources that are logarithmic in the dimensionality of quantum states. For pure quantum states, these resources can also be logarithmic in the number of quantum states used for training the machine-learning algorithm. This makes these algorithms potentially applicable to big quantum data applications.
收起
摘要 :
Quantum Machine Learning (QML) is one of the core research fields in the larger paradigm of Quantum Computing (also known alternatively as Quantum Information). In recent years, researchers have taken deep interest in QML, given t...
展开
Quantum Machine Learning (QML) is one of the core research fields in the larger paradigm of Quantum Computing (also known alternatively as Quantum Information). In recent years, researchers have taken deep interest in QML, given the potential time and cost advantages that solutions to real-life problems using QML algorithms provide, in comparison to their classical (or digital) machine learning equivalents. This is still a very new and exciting area of research with new algorithms and their uses being developed almost every other day. Deep research interest in this area has picked up only in the past 5-6 years. Given the background, this paper focuses on studying Scopus and Web of Science databases for the past 6 years (2014-2019) to identify various publication trends in the areas of Quantum Machine Learning. The authors have done an in-depth study of the Scopus and Web of Science publication data pertaining to this area and have come up with interesting insights. The survey covers 276 publications in Scopus and 154 publications in Web of Science. From the Scopus database, it is found that there has been a consistent growth in the number of publications in this period. Four research areas, namely, Physics, Astronomy, Computer Science, and Mathematics, have contributed 68.1% of the research publications. The USA leads the top 10 countries with nearly half (49.2%) of the research publications. A total of 148 patents have been published with 94 of these being published in the last four years (2016-2019). This essentially translates to one patent for every two publications. The Web of Science database, though bringing out 154 publications in the period, shows similar trends across the metrics. We have carried out a comparative study of some of the metrics in Scopus and Web of Science databases. Overall the study identifies the top 10 Institutions, authors, and research journals.
收起
原文获取