Predicting impurity spectral functions using machine learning

Abstract

The Anderson Impurity Model (AIM) is a canonical model of quantum many-body physics. Here we investigate whether machine learning models, both neural networks (NN) and kernel ridge regression (KRR), can accurately predict the AIM spectral function in all of its regimes, from empty orbital, to mixed valence, to Kondo. To tackle this question, we construct two large spectral databases containing approximately 410k and 600k spectral functions of the single-channel impurity problem. We show that the NN models can accurately predict the AIM spectral function in all of its regimes, with point-wise mean absolute errors down to 0.003 in normalized units. We find that the trained NN models outperform models based on KRR and enjoy a speedup on the order of $10^5$ over traditional AIM solvers. The required size of the training set of our model can be significantly reduced using furthest point sampling in the AIM parameter space, which is important for generalizing our method to more complicated multi-channel impurity problems of relevance to predicting the properties of real materials.

arXiv comment31 pages, 27 figures, 5 tables

References in corpus (23)

• Scikit-learn: Machine Learning in Python
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Kondo Physics in a Single Electron Transistor
• A Tunable Kondo Effect in Quantum Dots
• Continuous-time Monte Carlo methods for quantum impurity models
• The numerical renormalization group method for quantum impurity systems
• Theory of Intertwined Orders in High Temperature Superconductors
• Nonequilibrium dynamical mean-field theory and its applications
• Learning phase transitions by confusion
• Heavy Fermions and Quantum Phase Transitions
• Real-time dynamics in Quantum Impurity Systems: A Time-dependent Numerical Renormalization Group Approach
• One-dimensional Topological Edge States of Bismuth Bilayers
• Sum-rule Conserving Spectral Functions from the Numerical Renormalization Group
• Numerical Renormalization Group Calculations for the Self-energy of the impurity Anderson model
• Machine learning for many-body physics: The case of the Anderson impurity model
• DMFT+NRG study of spin-orbital separation in a three-band Hund's metal
• Volume-Collapse Transitions in the Rare Earth Metals
• Kondo decoherence: finding the right spin model for iron impurities in gold and silver
• Doublon-holon origin of the subpeaks at the Hubbard band edges
• A machine learning approach to dynamical properties of quantum many-body systems
• Unusual electronic and vibrational properties in the colossal thermopower material FeSb${}_2$
• Anisotropic Correlated Electronic Structure of Colossal Thermopower Marcasite FeSb${}_2$
• Machine learning effective models for quantum systems

Cited by in corpus (10)

• Machine learning of Kondo physics using variational autoencoders and symbolic regression
• Data-Driven Dynamical Mean-Field Theory: an error-correction approach to solve the quantum many-body problem using machine learning
• Equivariant neural network for Green's functions of molecules and materials
• Neural-network-supported basis optimizer for the configuration interaction problem in quantum many-body clusters: Feasibility study and numerical proof
• Machine-learning the spectral function of a hole in a quantum antiferromagnet
• Exploring fundamental laws of classical mechanics via predicting the orbits of planets based on neural networks
• Physically interpretable approximations of many-body spectral functions
• Hamiltonian parameter inference from resonant inelastic x-ray scattering with active learning
• Machine Learning Green's Functions of Strongly Correlated Hubbard Models
• Linear Foundation Model for Quantum Embedding: Data-Driven Compression of the Ghost Gutzwiller Variational Space