Unsupervised Learning of Effective Quantum Impurity Models

TL;DR

This paper introduces a non-perturbative, unsupervised machine learning method based on renormalization group principles to accurately derive low-energy effective quantum impurity models, improving simulations of complex correlated systems.

Contribution

It presents a novel, general, and systematically improvable machine learning approach for deriving effective impurity models, surpassing traditional perturbative methods.

Findings

Accurately reproduces low-energy physics of the Anderson impurity model

Demonstrates flexibility and systematic improvability of the method

Provides a pathway for studying more complex impurity models

Abstract

Generalized quantum impurity models -- which feature a few localized and strongly-correlated degrees of freedom coupled to itinerant conduction electrons -- describe diverse physical systems, from magnetic moments in metals to nanoelectronics quantum devices such as quantum dots or single-molecule transistors. Correlated materials can also be understood as self-consistent impurity models through dynamical mean field theory. Accurate simulation of such models is challenging, especially at low temperatures, due to many-body effects from electronic interactions, resulting in strong renormalization. In particular, the interplay between local impurity complexity and Kondo physics is highly nontrivial. A common approach, which we further develop in this work, is to consider instead a simpler effective impurity model that still captures the low-energy physics of interest. The mapping from bare to effective model is typically done perturbatively, but even this can be difficult for complex systems, and the resulting effective model parameters can anyway be quite inaccurate. Here we develop a non-perturbative, unsupervised machine learning approach to systematically obtain low-energy effective impurity-type models, based on the renormalization group framework. The method is shown to be general and flexible, as well as accurate and systematically improvable. We benchmark the method against exact results for the Anderson impurity model, and provide an outlook for more complex models beyond the reach of existing methods.