Deep learning Local Reduced Density Matrices for Many-body Hamiltonian Estimation

Abstract

Human experts cannot efficiently access the physical information of quantum many-body states by simply "reading" the coefficients, but have to reply on the previous knowledge such as order parameters and quantum measurements. In this work, we demonstrate that convolutional neural network (CNN) can learn from the coefficients of local reduced density matrices to estimate the physical parameters of the many-body Hamiltonians, such as coupling strengths and magnetic fields, provided the states as the ground states. We propose QubismNet that consists of two main parts: the Qubism map that visualizes the ground states (or the purified reduced density matrices) as images, and a CNN that maps the images to the target physical parameters. By assuming certain constraints on the training set for the sake of balance, QubismNet exhibits impressive powers of learning and generalization on several quantum spin models. While the training samples are restricted to the states from certain ranges of the parameters, QubismNet can accurately estimate the parameters of the states beyond such training regions. For instance, our results show that QubismNet can estimate the magnetic fields near the critical point by learning from the states away from the critical vicinity. Our work illuminates a data-driven way to infer the Hamiltonians that give the designed ground states, and therefore would benefit the existing and future generalizations of quantum technologies such as Hamiltonian-based quantum simulations and state tomography.

arXiv comment11 pages, 8 figures

References in corpus (37)

• Improving neural networks by preventing co-adaptation of feature detectors
• Quantum Simulation
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems
• Quantum Circuit Learning
• Machine learning phases of matter
• Learning phase transitions by confusion
• Matrix product states represent ground states faithfully
• Quantum Annealing and Analog Quantum Computation
• Efficient quantum state tomography
• Discovering Phase Transitions with Unsupervised Learning
• Tensor networks for complex quantum systems
• Machine Learning Topological States
• Differentiable Learning of Quantum Circuit Born Machine
• A Review of Object Detection Models based on Convolutional Neural Network
• Identifying topological order through unsupervised machine learning
• Machine Learning Topological Invariants with Neural Networks
• Soundness and completeness of quantum root-mean-square errors
• Symmetries and many-body excited states with neural-network quantum states
• Efficient tomography of a quantum many-body system
• An introduction to integrable techniques for one-dimensional quantum systems
• Identifying Quantum Phase Transitions using Artificial Neural Networks on Experimental Data
• Training of Quantum Circuits on a Hybrid Quantum Computer
• Neural-Network Quantum States, String-Bond States, and Chiral Topological States
• Accelerated search and design of stretchable graphene kirigami using machine learning
• Lecture Notes of Tensor Network Contractions
• Unsupervised machine learning and band topology
• Hiding Quiet Solutions in Random Constraint Satisfaction Problems
• Artificial Neural Network Approach to the Analytic Continuation Problem
• Transfer Matrices and Excitations with Matrix Product States
• Forward and Inverse Design of Kirigami via Supervised Autoencoder
• Quantum Spin Liquid in a Breathing Kagome Lattice
• Exascale Deep Learning for Scientific Inverse Problems
• Stability of the spin-$1/2$ kagome ground state with breathing anisotropy
• Qubism: self-similar visualization of many-body wavefunctions

Cited by in corpus (7)

• Self-Correcting Quantum Many-Body Control using Reinforcement Learning with Tensor Networks
• From observations to complexity of quantum states via unsupervised learning
• Compressing Neural Networks Using Tensor Networks with Exponentially Fewer Variational Parameters
• Supervised Training of Neural-Network Quantum States for the Next Nearest Neighbor Ising model
• Learning quantum dissipation by the neural ordinary differential equation
• Quantum compiling with a variational instruction set for accurate and fast quantum computing
• Exploring fundamental laws of classical mechanics via predicting the orbits of planets based on neural networks