Machine Learning by Unitary Tensor Network of Hierarchical Tree Structure

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.

6 pages, 4 figures

References in corpus (11)

• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• A class of quantum many-body states that can be efficiently simulated
• Towards Quantum Machine Learning with Tensor Networks
• Experimental Realization of Quantum Artificial Intelligence
• Unsupervised Generative Modeling Using Matrix Product States
• Tensor Networks for Dimensionality Reduction and Large-Scale Optimizations. Part 2 Applications and Future Perspectives
• Neural-Network Quantum States, String-Bond States, and Chiral Topological States
• Advances on Tensor Network Theory: Symmetries, Fermions, Entanglement, and Holography
• Machine Learning Spatial Geometry from Entanglement Features
• Holography as deep learning
• Deep learning and the renormalization group

Cited by in corpus (57)

• Parameterized quantum circuits as machine learning models
• Transfer learning in hybrid classical-quantum neural networks
• Machine Learning for Quantum Matter
• Variational Quantum Reinforcement Learning via Evolutionary Optimization
• An end-to-end trainable hybrid classical-quantum classifier
• Supervised Learning with Projected Entangled Pair States
• The Presence and Absence of Barren Plateaus in Tensor-network Based Machine Learning
• Generative Tensor Network Classification Model for Supervised Machine Learning
• Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions
• Riemannian geometry and automatic differentiation for optimization problems of quantum physics and quantum technologies
• Tensor networks for quantum machine learning
• Tensor networks for unsupervised machine learning
• Classical versus Quantum: comparing Tensor Network-based Quantum Circuits on LHC data
• Tree tensor network classifiers for machine learning: from quantum-inspired to quantum-assisted
• Gauging tensor networks with belief propagation
• Generative machine learning with tensor networks: benchmarks on near-term quantum computers
• Cost function embedding and dataset encoding for machine learning with parameterized quantum circuits
• QGOpt: Riemannian optimization for quantum technologies
• Compressing deep neural networks by matrix product operators
• Entanglement-guided architectures of machine learning by quantum tensor network
• Automatically Differentiable Quantum Circuit for Many-qubit State Preparation
• Modeling Sequences with Quantum States: A Look Under the Hood
• Quantum Machine Learning Tensor Network States
• Experimental realization of a quantum image classifier via tensor-network-based machine learning
• Dequantizing quantum machine learning models using tensor networks
• Mutual Information Scaling for Tensor Network Machine Learning
• Quantum Compressed Sensing with Unsupervised Tensor-Network Machine Learning
• Tangent-Space Gradient Optimization of Tensor Network for Machine Learning
• Unsupervised Recognition of Informative Features via Tensor Network Machine Learning and Quantum Entanglement Variations
• Learning Effective Spin Hamiltonian of Quantum Magnet
• Machine learning with tree tensor networks, CP rank constraints, and tensor dropout
• NoRA: A Tensor Network Ansatz for Volume-Law Entangled Equilibrium States of Highly Connected Hamiltonians
• Explainable Natural Language Processing with Matrix Product States
• Algorithms for Tensor Network Contraction Ordering
• Neural network enhanced hybrid quantum many-body dynamical distributions
• Privacy-preserving machine learning with tensor networks
• Stack operation of tensor networks
• Interaction Decompositions for Tensor Network Regression
• Residual Matrix Product State for Machine Learning
• The resource theory of tensor networks
• Deep tensor networks with matrix product operators
• Non-parametric Semi-Supervised Learning in Many-body Hilbert Space with Rescaled Logarithmic Fidelity
• Exploring explicit coarse-grained structure in artificial neural networks
• Cons-training Tensor Networks: Embedding and Optimization Over Discrete Linear Constraints
• Survey on Computational Applications of Tensor Network Simulations
• TensorKrowch: Smooth integration of tensor networks in machine learning
• Universal scaling laws in quantum-probabilistic machine learning by tensor network towards interpreting representation and generalization powers
• Quantum tensor network algorithms for evaluation of spectral functions on quantum computers
• Tensor-Networks-based Learning of Probabilistic Cellular Automata Dynamics
• Scaling of contraction costs for entanglement renormalization algorithms including tensor Trotterization and variational Monte Carlo
• Tensor Network Efficiently Representing Schmidt Decomposition of Quantum Many-Body States
• Entanglement renormalization for quantum fields with boundaries and defects
• Harnessing Quantum Dynamics for Robust and Scalable Quantum Extreme Learning Machines
• Explaining Anomalies with Tensor Networks
• Quantum Machine Learning with SQUID
• Trainable Quantum Neural Network for Multiclass Image Classification with the Power of Pre-trained Tree Tensor Networks
• Tensorization of neural networks for improved privacy and interpretability