Supervised Learning with Projected Entangled Pair States

Abstract

Tensor networks, a model that originated from quantum physics, has been gradually generalized as efficient models in machine learning in recent years. However, in order to achieve exact contraction, only tree-like tensor networks such as the matrix product states and tree tensor networks have been considered, even for modeling two-dimensional data such as images. In this work, we construct supervised learning models for images using the projected entangled pair states (PEPS), a two-dimensional tensor network having a similar structure prior to natural images. Our approach first performs a feature map, which transforms the image data to a product state on a grid, then contracts the product state to a PEPS with trainable parameters to predict image labels. The tensor elements of PEPS are trained by minimizing differences between training labels and predicted labels. The proposed model is evaluated on image classifications using the MNIST and the Fashion-MNIST datasets. We show that our model is significantly superior to existing models using tree-like tensor networks. Moreover, using the same input features, our method performs as well as the multilayer perceptron classifier, but with much fewer parameters and is more stable. Our results shed light on potential applications of two-dimensional tensor network models in machine learning.

7 pages, 4 figures, 1 table

References in corpus (34)

• Adam: A Method for Stochastic Optimization
• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Area laws for the entanglement entropy - a review
• A Practical Introduction to Tensor Networks: Matrix Product States and Projected Entangled Pair States
• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Classical simulation of infinite-size quantum lattice systems in two spatial dimensions
• Area laws in quantum systems: mutual information and correlations
• Tensor renormalization group approach to 2D classical lattice models
• Tensorizing Neural Networks
• Criticality, the area law, and the computational power of PEPS
• Training Deep Nets with Sublinear Memory Cost
• Quantum Entanglement in Neural Network States
• Why does deep and cheap learning work so well?
• Efficient Representation of Quantum Many-body States with Deep Neural Networks
• The computational complexity of PEPS
• Differentiable Learning of Quantum Circuit Born Machine
• Unsupervised Generative Modeling Using Matrix Product States
• Equivalence of restricted Boltzmann machines and tensor network states
• Differentiable Programming Tensor Networks
• Algorithms for finite Projected Entangled Pair States
• Quantum Entanglement in Deep Learning Architectures
• Advances on Tensor Network Theory: Symmetries, Fermions, Entanglement, and Holography
• Unifying Projected Entangled Pair States contractions
• Tree Tensor Networks for Generative Modeling
• Machine Learning by Unitary Tensor Network of Hierarchical Tree Structure
• Learning Relevant Features of Data with Multi-scale Tensor Networks
• Information Perspective to Probabilistic Modeling: Boltzmann Machines versus Born Machines
• Supervised Learning with Quantum-Inspired Tensor Networks
• TensorNetwork: A Library for Physics and Machine Learning
• TensorNetwork for Machine Learning
• Matrix Product Operators for Sequence to Sequence Learning
• An efficient quantum algorithm for generative machine learning
• Entanglement Entropy of Target Functions for Image Classification and Convolutional Neural Network

Cited by in corpus (32)

• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Accurate simulation for finite projected entangled pair states in two dimensions
• Tensor networks for unsupervised machine learning
• From Tensor Network Quantum States to Tensorial Recurrent Neural Networks
• Quantum reservoir computation utilising scale-free networks
• Tensor networks for interpretable and efficient quantum-inspired machine learning
• Mutual Information Scaling for Tensor Network Machine Learning
• Accurate Simulation of the Hubbard Model with Finite Fermionic Projected Entangled Pair States
• Unsupervised Recognition of Informative Features via Tensor Network Machine Learning and Quantum Entanglement Variations
• Differentiable Programming of Isometric Tensor Networks
• Quantum-inspired event reconstruction with Tensor Networks: Matrix Product States
• Machine learning with tree tensor networks, CP rank constraints, and tensor dropout
• Stack operation of tensor networks
• Privacy-preserving machine learning with tensor networks
• Residual Matrix Product State for Machine Learning
• Dual-isometric Projected Entangled Pair States
• The resource theory of tensor networks
• Compressing Neural Networks Using Tensor Networks with Exponentially Fewer Variational Parameters
• Non-parametric Semi-Supervised Learning in Many-body Hilbert Space with Rescaled Logarithmic Fidelity
• Deep tensor networks with matrix product operators
• Tensor networks and efficient descriptions of classical data
• Infinitely Wide Tensor Networks as Gaussian Process
• Cons-training Tensor Networks: Embedding and Optimization Over Discrete Linear Constraints
• A Matrix Product State Model for Simultaneous Classification and Generation
• Exploring explicit coarse-grained structure in artificial neural networks
• Lower and Upper Bounds on the VC-Dimension of Tensor Network Models
• Quantum Deformed Neural Networks
• Tensor-Networks-based Learning of Probabilistic Cellular Automata Dynamics
• Scaling of contraction costs for entanglement renormalization algorithms including tensor Trotterization and variational Monte Carlo
• Quantum max-flow in the bridge graph
• Network Dynamics-Based Framework for Understanding Deep Neural Networks
• Using matrix-product states for time-series machine learning