Using matrix-product states for time-series machine learning

Abstract

Matrix-product states (MPS) have proven to be a versatile ansatz for modeling quantum many-body physics. For many applications, and particularly in one-dimension, they capture relevant quantum correlations in many-body wavefunctions while remaining tractable to store and manipulate on a classical computer. This has motivated researchers to also apply the MPS ansatz to machine learning (ML) problems where capturing complex correlations in datasets is also a key requirement. Here, we develop and apply an MPS-based algorithm, MPSTime, for learning a joint probability distribution underlying an observed time-series dataset, and show how it can be used to tackle important time-series ML problems, including classification and imputation. MPSTime can efficiently learn complicated time-series probability distributions directly from data, requires only moderate maximum MPS bond dimension ${\chi }_{\max }$, with values for our applications ranging between ${\chi }_{\max }=20-160$, and can be trained for both classification and imputation tasks under a single logarithmic loss function. Using synthetic and publicly available real-world datasets, spanning applications in medicine, energy, and astronomy, we demonstrate performance competitive with state-of-the-art ML approaches, but with the key advantage of encoding the full joint probability distribution learned from the data, which is useful for analyzing and interpreting its underlying structure. This manuscript is supplemented with the release of a publicly available code package MPSTime that implements our approach. The effectiveness of the MPS-based ansatz for capturing complex correlation structures in time-series data makes it a powerful foundation for tackling challenging time-series analysis problems across science, industry, and medicine.

31 pages, 14 figures

References in corpus (36)

• The density-matrix renormalization group in the age of matrix product states
• Efficient classical simulation of slightly entangled quantum computations
• A Practical Introduction to Tensor Networks: Matrix Product States and Projected Entangled Pair States
• Efficient simulation of one-dimensional quantum many-body systems
• InceptionTime: Finding AlexNet for Time Series Classification
• Continuous-variable optical quantum state tomography
• Classical simulation of infinite-size quantum lattice systems in one spatial dimension
• Matrix Product States and Projected Entangled Pair States: Concepts, Symmetries, and Theorems
• Matrix product states represent ground states faithfully
• Highly comparative time-series analysis: The empirical structure of time series and their methods
• The computational complexity of PEPS
• Unsupervised Generative Modeling Using Matrix Product States
• Photonic Quantum Circuits with Time Delays
• Bake off redux: a review and experimental evaluation of recent time series classification algorithms
• Dynamical Variational Autoencoders: A Comprehensive Review
• Efficient tensor network simulation of IBM's Eagle kicked Ising experiment
• Perfect Sampling with Unitary Tensor Networks
• Matrix product states for critical spin chains: finite size scaling versus finite entanglement scaling
• Impact on asteroseismic analyses of regular gaps in Kepler data
• Tree Tensor Networks for Generative Modeling
• Dynamics of many-body photon bound states in chiral waveguide QED
• Supervised Learning with Projected Entangled Pair States
• Simulating quantum light propagation through atomic ensembles using matrix product states
• Contracting projected entangled pair states is average-case hard
• Tensor networks for unsupervised machine learning
• Entanglement-guided architectures of machine learning by quantum tensor network
• Chiral Light--Matter Interaction Beyond the Rotating-Wave Approximation
• 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
• Quantum-inspired event reconstruction with Tensor Networks: Matrix Product States
• Tensor Networks for Explainable Machine Learning in Cybersecurity
• Positive unlabeled learning with tensor networks
• Random insights into the complexity of two-dimensional tensor network calculations
• Tensor network to learn the wavefunction of data
• A Matrix Product State Model for Simultaneous Classification and Generation