Tensor networks for unsupervised machine learning

Abstract

Modeling the joint distribution of high-dimensional data is a central task in unsupervised machine learning. In recent years, many interests have been attracted to developing learning models based on tensor networks, which have the advantages of a principle understanding of the expressive power using entanglement properties, and as a bridge connecting classical computation and quantum computation. Despite the great potential, however, existing tensor network models for unsupervised machine learning only work as a proof of principle, as their performance is much worse than the standard models such as restricted Boltzmann machines and neural networks. In this Letter, we present autoregressive matrix product states (AMPS), a tensor network model combining matrix product states from quantum many-body physics and autoregressive modeling from machine learning. Our model enjoys the exact calculation of normalized probability and unbiased sampling. We demonstrate the performance of our model using two applications, generative modeling on synthetic and real-world data, and reinforcement learning in statistical physics. Using extensive numerical experiments, we show that the proposed model significantly outperforms the existing tensor network models and the restricted Boltzmann machines, and is competitive with state-of-the-art neural network models.

arXiv commentv2

References in corpus (20)

• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Characterizing Quantum Supremacy in Near-Term Devices
• Classical simulation of quantum many-body systems with a tree tensor network
• Low-Rank Tensor Networks for Dimensionality Reduction and Large-Scale Optimization Problems: Perspectives and Challenges PART 1
• MADE: Masked Autoencoder for Distribution Estimation
• Unsupervised Generative Modeling Using Matrix Product States
• Deep autoregressive models for the efficient variational simulation of many-body quantum systems
• Solving Statistical Mechanics Using Variational Autoregressive Networks
• Neural-Network Quantum States, String-Bond States, and Chiral Topological States
• Tree Tensor Networks for Generative Modeling
• Asymptotically unbiased estimation of physical observables with neural samplers
• 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
• Unifying Neural-network Quantum States and Correlator Product States via Tensor Networks
• Contracting Arbitrary Tensor Networks: General Approximate Algorithm and Applications in Graphical Models and Quantum Circuit Simulations
• Supervised Learning with Projected Entangled Pair States
• Simulating the Sycamore quantum supremacy circuits
• Boltzmann machines as two-dimensional tensor networks
• Solving Statistical Mechanics on Sparse Graphs with Feedback Set Variational Autoregressive Networks

Cited by in corpus (15)

• From Tensor Network Quantum States to Tensorial Recurrent Neural Networks
• Dequantizing quantum machine learning models using tensor networks
• Tensor networks for quantum computing
• Machine learning with tree tensor networks, CP rank constraints, and tensor dropout
• Tensor networks and efficient descriptions of classical data
• A Matrix Product State Model for Simultaneous Classification and Generation
• Efficient Optimization of Variational Autoregressive Networks with Natural Gradient
• Tensor-Networks-based Learning of Probabilistic Cellular Automata Dynamics
• Universal scaling laws in quantum-probabilistic machine learning by tensor network towards interpreting representation and generalization powers
• Optimal Qubit Mapping Search for Encoding Classical Data into Matrix Product State Representation with Minimal Loss
• The Cytnx Library for Tensor Networks
• A Synthesis of Hidden Subgroup Quantum Algorithms and Quantum Chemical Dynamics
• Harnessing Quantum Dynamics for Robust and Scalable Quantum Extreme Learning Machines
• Time series generation for option pricing on quantum computers using tensor network
• Using matrix-product states for time-series machine learning