Effects of Noise on the Overparametrization of Quantum Neural Networks

By Diego García-Martín et al

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

Table of Contents

I. Introduction
II. Framework
A. Quantum Neural Networks
B. Dynamical Lie Algebra, Quantum Fisher Information, and Overparametrization
C. Quantum Noise Preliminaries

Summary

The paper discusses the effects of noise on the overparametrization phenomenon in Quantum Neural Networks (QNNs). It explores how noise can impact the exploration of state space directions, potentially turning an overparametrized QNN into an underparametrized one. The study shows that noise can enable new directions in state space but also suppress sensitivity to parameter updates. The presence of hardware noise alters the understanding of QNNs, and the paper provides insights on how noise affects the capacity measures of QNNs. The concepts of Dynamical Lie Algebra, Quantum Fisher Information, and Quantum noise are discussed in the context of QNNs.