Optimization by VarQITE on Adaptive Variational Quantum Kolmogorov-Arnold Network

TL;DR

This paper introduces a novel application of quantum imaginary time evolution (QITE) for quantum machine learning, demonstrating improved accuracy over quantum neural networks in function fitting and classification tasks.

Contribution

It presents the first method to apply QITE to quantum machine learning, expanding its utility beyond traditional ground state problems.

Findings

QITE-based method outperforms quantum neural networks in certain tasks

Successfully applied to function fitting and classification on a 2-D plane

Potential to be a milestone for integrating QITE into quantum machine learning

Abstract

Quantum imaginary time evolution (QITE) is a powerful method to derive the ground states of the systems. Only the damping of quantum states leads it; hence, reaching the ground state is guaranteed by nature without any external manipulation. Numerous QITE methods by many groups are used to improve speed and accuracy, derive excited states, and solve combined optimization problems. However, the QITE methods have not been used for quantum machine learning to predict the ideal values for multiple input values. Therefore, we propose a method for applying QITE methods for quantum machine learning and demonstrate fitting problems of elementary functions and classification problems on a 2-D plane. As a result, we confirmed that our method was more accurate than a quantum neural network in solving some problems. Our method can be used for other quantum machine learning algorithms; hence, it may be the milestone for applying QITE to quantum machine learning.