An Algorithm for Fast Supervised Learning in Variational Circuits through Simultaneous Processing of Multiple Samples

TL;DR

This paper introduces a quantum algorithm that accelerates the training of variational classifiers by processing multiple samples simultaneously, reducing training complexity from linear to logarithmic in dataset size.

Contribution

The paper presents a novel quantum algorithm that leverages qRAM and Swap-test circuits to significantly speed up variational classifier training, adaptable to various ansatzes.

Findings

Training cost reduced to O(logN) from O(N)

Algorithm applicable to any variational ansatz

Potential extension to multi-class classification

Abstract

We propose a novel algorithm for fast training of variational classifiers by processing multiple samples parallelly. The algorithm can be adapted for any ansatz used in the variational circuit. The presented algorithm utilizes qRAM and other quantum circuits in the forward pass. Further, instead of the usual practice of computing the loss classically, we calculate the loss using a Swap-test circuit. The algorithm thus brings down the training cost of a variational classifier to O(logN)from the usual O(N)when training on a dataset of N samples. Although we discuss only binary classification in the paper, the algorithm can be easily generalized to multi-class classification.