Saved in:
Bibliographic Details
Main Authors: Wen, Zujin, Huang, Jin-Long, Dahlsten, Oscar
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.22016
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908425928048640
author Wen, Zujin
Huang, Jin-Long
Dahlsten, Oscar
author_facet Wen, Zujin
Huang, Jin-Long
Dahlsten, Oscar
contents We design a quantum version of neural networks with sinusoidal activation functions and compare its performance to the classical case. We create a general quantum sine circuit implementing a discretised sinusoidal activation function. Along the way, we define a classical discrete sinusoidal neural network. We build a quantum optimization algorithm around the quantum sine circuit, combining quantum search and phase estimation. This algorithm is guaranteed to find the weights with global minimum loss on the training data. We give a computational complexity analysis and demonstrate the algorithm in an example. We compare the performance with that of the standard gradient descent training method for classical sinusoidal neural networks. We show that (i) the standard classical training method typically leads to bad local minima in terms of mean squared error on test data and (ii) the weights that perform best on the training data generalise well to the test data. Points (i) and (ii) motivate using the quantum training algorithm, which is guaranteed to find the best weights on the training data.
format Preprint
id arxiv_https___arxiv_org_abs_2410_22016
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Quantum Sinusoidal Neural Networks
Wen, Zujin
Huang, Jin-Long
Dahlsten, Oscar
Quantum Physics
We design a quantum version of neural networks with sinusoidal activation functions and compare its performance to the classical case. We create a general quantum sine circuit implementing a discretised sinusoidal activation function. Along the way, we define a classical discrete sinusoidal neural network. We build a quantum optimization algorithm around the quantum sine circuit, combining quantum search and phase estimation. This algorithm is guaranteed to find the weights with global minimum loss on the training data. We give a computational complexity analysis and demonstrate the algorithm in an example. We compare the performance with that of the standard gradient descent training method for classical sinusoidal neural networks. We show that (i) the standard classical training method typically leads to bad local minima in terms of mean squared error on test data and (ii) the weights that perform best on the training data generalise well to the test data. Points (i) and (ii) motivate using the quantum training algorithm, which is guaranteed to find the best weights on the training data.
title Quantum Sinusoidal Neural Networks
topic Quantum Physics
url https://arxiv.org/abs/2410.22016