Once developed for quantum theory, tensor networks (TNs) have been established as a successful machine learning (ML) paradigm. Now, they have been ported back to the quantum realm in the emerging field of quantum ML to assess problems that classical computers are unable to solve efficiently. Their nature at the interface between physics and ML makes TNs easily deployable on quantum computers. In this review article, we shed light on one of the major architectures considered to be predestined for variational quantum ML. In particular, we discuss how layouts like matrix product state, projected entangled pair states, tree tensor networks and multi-scale entanglement renormalization ansatz can be mapped to a quantum computer, how they can be used for ML and data encoding and which implementation techniques improve their performance.
Once developed for quantum theory, tensor networks (TNs) have been established as a successful machine learning (ML) paradigm. Now, they have been ported back to the quantum realm in the emerging field of quantum ML to assess problems that classical computers are unable to solve efficiently. Their nature at the interface between physics and ML makes TNs easily deployable on quantum computers. In this review article, we shed light on one of the major architectures considered to be predestined for variational quantum ML. In particular, we discuss how layouts like matrix product state, projected entangled pair states, tree tensor networks and multi-scale entanglement renormalization ansatz can be mapped to a quantum computer, how they can be used for ML and data encoding and which implementation techniques improve their performance.
Abstract There have been numerous quantum neural networks reported, but they struggle to match traditional neural networks in accuracy. Given the huge improvement of the neural network models’ accuracy by two-dimensional tensor network (TN) states in classical tensor network machine learning (TNML), it is promising to explore whether its application in quantum machine learning can extend the performance boundary of the models. Here, we transform two-dimensional TNs into quantum circuits for supervised learning. Specifically, we encode two-dimensional TNs into quantum circuits through rigorous mathematical proofs for constructing model ansätze, including string-bond states, entangled-plaquette states and isometric TN states. In addition, we propose adaptive data encoding methods and combine with TNs. We construct a tensor-network-inspired quantum circuit (TNQC) supervised learning framework for transferring TNML from classical to quantum, and build several novel two-dimensional TN-inspired quantum classifiers based on this framework. Finally, we propose a parallel quantum machine learning method for multi-class classification to construct 2D TNQC-based multi-class classifiers. Classical simulation results on the MNIST benchmark dataset show that our proposed models achieve the state-of-the-art accuracy performance, significantly outperforming other quantum classifiers on both binary and multi-class classification tasks, and beat simple convolutional classifiers on a fair track with identical inputs. The noise resilience of the models makes them successfully run and work in a real quantum computer.
Over the past few years a number of proofs have emerged revealing the many connections between the methods used in quantum-many body physics and those in machine learning. In particular, much attention has been given to tensor networks (TNs) and deep learning architectures which exhibit striking similarities. Finding those similarities has helped us gain a better understanding on why deep learning architectures have so much expressive efficiency. Recently, machine learning techniques have been used to approximate many-body physics problems. Conversely, TNs have been used for machine learning tasks. For example, state-of-the-art research has used one-dimensional TNs to solve image recognition problems with very limited scalability. The scalability problem, however, can be overcomed by using a two-dimensional hierarchical TNs and a training algorithm derived from the multipartite entanglement renormalization ansatz (MERA). Here we give further analysis of the inner structural resemblance between such hierarchical tensor networks (TNs) and pretrained convolutional neural networks (CNNs): this was achieved by analyzing their abstraction power layer by layer.
Tensor networks (TNs) have become one of the most essential building blocks for various fields of theoretical physics such as condensed matter theory, statistical mechanics, quantum information, and quantum gravity. This review provides a unified description of a series of developments in the TN from the statistical mechanics side. In particular, we begin with the variational principle for the transfer matrix of the 2D Ising model, which naturally leads us to the matrix product state (MPS) and the corner transfer matrix (CTM). We then explain how the CTM can be evolved to such MPS-based approaches as density matrix renormalization group (DMRG) and infinite time-evolved block decimation. We also elucidate that the finite-size DMRG played an intrinsic role for incorporating various quantum information concepts in subsequent developments in the TN. After surveying higher-dimensional generalizations like tensor product states or projected entangled pair states, we describe tensor renormalization groups (TRGs), which are a fusion of TNs and Kadanoff-Wilson type real-space renormalization groups, focusing on their fixed point structures. We then discuss how the difficulty in TRGs for critical systems can be overcome in the tensor network renormalization and the multi-scale entanglement renormalization ansatz.
Abstract The resemblance between the methods used in quantum-many body physics and in machine learning has drawn considerable attention. In particular, tensor networks (TNs) and deep learning architectures bear striking similarities to the extent that TNs can be used for machine learning. Previous results used one-dimensional TNs in image recognition, showing limited scalability and flexibilities. In this work, we train two-dimensional hierarchical TNs to solve image recognition problems, using a training algorithm derived from the multi-scale entanglement renormalization ansatz. This approach introduces mathematical connections among quantum many-body physics, quantum information theory, and machine learning. While keeping the TN unitary in the training phase, TN states are defined, which encode classes of images into quantum many-body states. We study the quantum features of the TN states, including quantum entanglement and fidelity. We find these quantities could be properties that characterize the image classes, as well as the machine learning tasks.
Tensor networks were developed in the context of many-body physics as compressed representations of multiparticle quantum states. These representations mitigate the exponential complexity of many-body systems by capturing only the most relevant dependencies. Due to the formal similarity between quantum entanglement and statistical correlations, tensor networks have recently been integrated in machine learning, operating both as alternative learning architectures and as decompositions of components of neural networks. The expectation is that the theoretical understanding of tensor networks developed within quantum many-body physics leads to novel methods that offer advantages in terms of computational efficiency, explainability, or privacy. Here we review the use of tensor networks in the context of machine learning, providing a critical assessment of the state of the art, the potential advantages, and the challenges that must be overcome.
Abstract Machine learning is a promising application of quantum computing, but challenges remain for implementation today because near-term devices have a limited number of physical qubits and high error rates. Motivated by the usefulness of tensor networks for machine learning in the classical context, we propose quantum computing approaches to both discriminative and generative learning, with circuits based on tree and matrix product state tensor networks, that could already have benefits with such near-term devices. The result is a unified framework in which classical and quantum computing can benefit from the same theoretical and algorithmic developments, and the same model can be trained classically then transferred to the quantum setting for additional optimization. Tensor network circuits can also provide qubit-efficient schemes in which, depending on the architecture, the number of physical qubits required scales only logarithmically with, or independently of the input or output data sizes. We demonstrate our proposals with numerical experiments, training a discriminative model to perform handwriting recognition using a hybrid quantum-classical optimization procedure that could be carried out on quantum hardware today, and testing the noise resilience of the trained model.