Hierarchical quantum classifiers

By Edward Grant, Marcello Benedetti, Shuxiang Cao, Andrew Hallam, Joshua Lockhart, Vid Stojevic, Andrew G. Green, Simone Severini. npj Quantum Information, 2018

381 citations recorded by the source corpus.

Quantum circuits with hierarchical structure have been used to perform binary classification of classical data encoded in a quantum state. We demonstrate that more expressive circuits in the same family achieve better accuracy and can be used to classify highly entangled quantum states, for which there is no known efficient classical method. We compare performance for several different parameterizations on two classical machine learning datasets, Iris and MNIST, and on a synthetic dataset of quantum states. Finally, we demonstrate that performance is robust to noise and deploy an Iris dataset classifier on the ibmqx4 quantum computer. Quantum algorithms with hierarchical tensor network structures may provide an efficient approach to machine learning using quantum computers. Recent theoretical work has indicated that quantum algorithms could have an advantage over classical methods for the linear algebra computations involved in machine learning. At the same time, mathematical structures called tensor networks, with some similarities to neural networks, have been shown to represent quantum states and circuits that can be efficiently evaluated. Edward Grant from University College London and colleagues from the UK and China have shown how quantum algorithms based on two tensor network structures can be used to classify both classical and quantum data. If implemented on a large scale quantum computer, their approach may enable classification of two-dimensional images and entangled quantum data more efficiently than is possible with classical methods.

Subfields: Artificial Intelligence, Atomic and Molecular Physics, and Optics

View Paper: https://www.semanticscholar.org/paper/7c2271ddf761a70a5e5668cd999e9589d12f79c4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Atomic and Molecular Physics, and Optics

Citations in the corpus

An initialization strategy for addressing barren plateaus in parametrized quantum circuits (2019, methodology): “Resolving this issue is critical to the scalability of algorithms such as the variational quantum eigensolver (VQE) [2, 3] and quantum neural networks (QNNs) [4, 5, 6].”
Quantum computing with neutral atoms (2020, background): “New quantum applications for Machine learning are today rapidly developing [7,84], and several classifiers have been elaborated [85,86] as well as sampling methods of classically inaccessible systems.”
Parameterized quantum circuits as machine learning models (2019, methodology): “[30] Edward Grant, Marcello Benedetti, Shuxiang Cao, Andrew Hallam, Joshua Lockhart, Vid Stojevic, Andrew G.”
PennyLane: Automatic differentiation of hybrid quantum-classical computations (2018, background): “[12] Edward Grant, Marcello Benedetti, Shuxiang Cao, Andrew Hallam, Joshua Lockhart, Vid Stojevic, Andrew G Green, and Simone Severini, “Hierarchical quantum classifiers,” arXiv preprint (2018), arxiv:1804.”
General parameter-shift rules for quantum gradients (2021, background): “de tion [17, 18], and variational quantum machine learning [19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31].”
Robust data encodings for quantum classifiers (2020, methodology): “Motivated by such NISQ limitations, some recent authors [34, 35, 65, 66] have considered a “qubit encoding””
Absence of Barren Plateaus in Quantum Convolutional Neural Networks (2020, background): “in [55] under the name Hierarchical Quantum Classifier, which our work applies to as well.”
Quantum convolutional neural network for classical data classification (2021, methodology): “For instance, circuit 1 is used as the parametrized quantum circuit for training a tree tensor network (TTN) [29].”