Efficient MPS representations and quantum circuits from the Fourier modes of classical image data

Abstract

Machine learning tasks are an exciting application for quantum computers, as it has been proven that they can learn certain problems more efficiently than classical ones. Applying quantum machine learning algorithms to classical data can have many important applications, as qubits allow for dealing with exponentially more data than classical bits. However, preparing the corresponding quantum states usually requires an exponential number of gates and therefore may ruin any potential quantum speedups. Here, we show that classical data with a sufficiently quickly decaying Fourier spectrum after being mapped to a quantum state can be well-approximated by states with a small Schmidt rank (i.e., matrix-product states) and we derive explicit error bounds. These approximated states can, in turn, be prepared on a quantum computer with a linear number of nearest-neighbor two-qubit gates. We confirm our results numerically on a set of $1024\times 1024$-pixel images taken from the `Imagenette' and DIV2K datasets. Additionally, we consider different variational circuit ansätze and demonstrate numerically that one-dimensional sequential circuits achieve the same compression quality as more powerful ansätze.

17 pages, 9 figures (+ 15 pages, 5 figures appendix); additional numerical data, published version

References in corpus (74)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Quantum Computing in the NISQ era and beyond
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Quantum Machine Learning
• The density-matrix renormalization group in the age of matrix product states
• Area laws for the entanglement entropy - a review
• Supervised learning with quantum enhanced feature spaces
• Barren plateaus in quantum neural network training landscapes
• Quantum support vector machine for big data classification
• Quantum Convolutional Neural Networks
• An Area Law for One Dimensional Quantum Systems
• A class of quantum many-body states that can be efficiently simulated
• Matrix Product States and Projected Entangled Pair States: Concepts, Symmetries, and Theorems
• Matrix product states represent ground states faithfully
• The effect of data encoding on the expressive power of variational quantum machine learning models
• Power of data in quantum machine learning
• A rigorous and robust quantum speed-up in supervised machine learning
• Quantum Data Fitting
• Entropy scaling and simulability by Matrix Product States
• Sequential generation of entangled multi-qubit states
• Quantum-state preparation with universal gate decompositions
• Variational quantum algorithms for nonlinear problems
• Hierarchical quantum classifiers
• Towards Quantum Machine Learning with Tensor Networks
• Image compression and entanglement
• Quantum Circuits for Isometries
• Real- and imaginary-time evolution with compressed quantum circuits
• Circuit-Based Quantum Random Access Memory for Classical Data
• Encoding of Matrix Product States into Quantum Circuits of One- and Two-Qubit Gates
• Isometric Tensor Network States in Two Dimensions
• Sequential Generation of Matrix-Product States in Cavity QED
• A Quantum Inspired Approach to Exploit Turbulence Structures
• Crossing a topological phase transition with a quantum computer
• Parallel Quantum Simulation of Large Systems on Small Quantum Computers
• Pattern recognition on a quantum computer
• Preparation of matrix product states with log-depth quantum circuits
• Supervised Learning with Quantum-Inspired Tensor Networks
• Emergent irreversibility and entanglement spectrum statistics
• Learning Feynman Diagrams with Tensor Trains
• The Variational Power of Quantum Circuit Tensor Networks
• Fault tolerant resource estimation of quantum random-access memories
• On the Quantum versus Classical Learnability of Discrete Distributions
• Quasioptimality of maximum-volume cross interpolation of tensors
• Barren plateaus in quantum tensor network optimization
• Mid-circuit measurements on a single species neutral alkali atom quantum processor
• Parallel cross interpolation for high-precision calculation of high-dimensional integrals
• Quantum-inspired algorithms for multivariate analysis: from interpolation to partial differential equations
• Scaling of variational quantum circuit depth for condensed matter systems
• Entanglement vs. gap for one-dimensional spin systems
• TensorNetwork for Machine Learning
• Quantics Tensor Cross Interpolation for High-Resolution, Parsimonious Representations of Multivariate Functions in Physics and Beyond
• Constant-depth preparation of matrix product states with adaptive quantum circuits
• Multigrid Renormalization
• Efficient quantum algorithm for preparing molecular-system-like states on a quantum computer
• Riemannian geometry and automatic differentiation for optimization problems of quantum physics and quantum technologies
• Quantum Monte Carlo Integration: The Full Advantage in Minimal Circuit Depth
• Optimization search effort over the control landscapes for open quantum systems with Kraus-map evolution
• Quantum pixel representations and compression for $N$-dimensional images
• Irreversibility and Entanglement Spectrum Statistics in Quantum Circuits
• Absence of barren plateaus in finite local-depth circuits with long-range entanglement
• Model-Independent Learning of Quantum Phases of Matter with Quantum Convolutional Neural Networks
• Suppression of mid-circuit measurement crosstalk errors with micromotion
• Quantum state preparation protocol for encoding classical data into the amplitudes of a quantum information processing register's wave function
• QGOpt: Riemannian optimization for quantum technologies
• Matrix Product State for Higher-Order Tensor Compression and Classification
• Linear-depth quantum circuits for loading Fourier approximations of arbitrary functions
• Drastic Circuit Depth Reductions with Preserved Adversarial Robustness by Approximate Encoding for Quantum Machine Learning
• A super-polynomial quantum-classical separation for density modelling
• Entanglement Transitions in Unitary Circuit Games
• Time Evolution of Uniform Sequential Circuits
• Exponential separations between classical and quantum learners
• Classification of the Fashion-MNIST Dataset on a Quantum Computer
• Finite-Depth Preparation of Tensor Network States from Measurement
• Tensor Network Based Efficient Quantum Data Loading of Images

Cited by in corpus (6)

• The State Preparation of Multivariate Normal Distributions using Tree Tensor Network
• Typical Machine Learning Datasets as Low-Depth Quantum Circuits
• Preparation Circuits for Matrix Product States by Classical Variational Disentanglement
• Pseudospectral method for solving PDEs using Matrix Product States
• TTNOpt: Tree tensor network package for high-rank tensor compression
• Quantum Encoding of Structured Data with Matrix Product States