Seeking a quantum advantage for machine learning

Ewin Tang

Pranab Rai, Tegil J John, Josaiah Murfreal Dkhar

Ximing Wang

In today's data-driven society, the importance of data is ever-increasing. The ability to discern patterns and trends in this data allows us to make predictive and informed decisions. This quest for enhanced data analysis has fueled the evolution of machine learning. Quantum Machine Learning (QML) is an emerging field that seeks to amalgamate the power of quantum computing with machine learning. Despite the term quantum in its name, QML is a hybrid approach incorporating both classical and quantum processing. This fusion of classical and quantum processing enables QML to capitalize on the advantages of both. However, this integration also introduces a novel set of challenges. Here we will explore QML from computer science and physics perspectives. In the first part, the mathematical tools for machine learning and quantum theory are introduced. Then the second part will investigate how quantum computation enhances classical machine learning algorithms, with a demonstration of quantum advantages in computational complexity. The third part explores the learning of quantum models through classical optimization methods. We will find how the features of a quantum model can be learned by optimization, and how a quantum model can be trained to generate target stochastic data.

Esma Aı̈meur, Gilles Brassard A, Sébastien Gambs

Zhenghao Li, Hao Zhan, Shana H. Winston +11 more

Quantum machine learning seeks a computational advantage in data processing by evaluating functions of quantum states, such as their similarity, that can be classically intractable to compute. For quantum advantage to be possible, however, it is essential to bypass costly characterisation of individual data instances in favour of efficient, direct similarity evaluation. Here we demonstrate a sample-optimal, hardware-efficient protocol for estimating quantum similarity -- the state overlap -- using bosonic quantum interference. The sample complexity of this approach is independent of the system dimension and is information-theoretically optimal up to a constant factor. Experimentally, we implement the scheme on \emph{Prakash-1}, a quantum computing platform based on a fully programmable integrated photonic processor. By preparing and interfering qudit states on the chip to directly extract their overlap, we demonstrate classification and online learning of quantum data with high accuracy in realistic noisy experiments. Our results establish joint overlap measurements as a scalable pathway to efficient quantum data analysis and a practical building block for network-integrated quantum machine learning.

Jincheng Zhang

This paper investigates the potential of leveraging quantum self-organization principles to develop novel machine learning algorithms. The core claim is to design a machine learning approach that benefits from quantum self-organization, aiming to enhance data analysis and model prediction efficiency. Traditional machine learning relies on classical algorithms, and this research seeks to exploit quantum self-organization for a deeper understanding of data analysis and model prediction. The study outlines a proposed algorithm and discusses its potential advantages and challenges.