Efficient and Quantum-Adaptive Machine Learning with Fermion Neural Networks

By Pei-Lin Zheng et al

Published on Sept. 19, 2023

Read the original document by opening this link in a new tab.

Table of Contents

I. Introduction
II. FNN for Machine Learning
III. Classical Examples
IV. Examples: Topological Insulators
V. Examples: Strongly Correlated Systems
VI. Physical Insights and Interpretability from FNN Machine Learning

Summary

This paper introduces Fermion Neural Networks (FNNs) for efficient machine learning, focusing on quantum-adaptive approaches. FNNs leverage physical properties to achieve competitive performance in challenging machine-learning benchmarks. They are applied to quantum systems without preprocessing, determining topological phases and emergent charge orders. The paper demonstrates successful characterizations of topological phases and charge orders using FNN machine learning. Furthermore, FNNs offer insights into the vanishing gradient problem in deep neural networks and provide novel perspectives for interpretable machine learning in the quantum realm.