7–11 Dec 2026
The University of Sydney
Australia/Sydney timezone
AIP Congress 2026

Machine learning of quantum data using optimal similarity measurements

Not scheduled
20m
Belinda Hutchinson Building (The University of Sydney )

Belinda Hutchinson Building

The University of Sydney

Abercrombie St & Codrington St NSW 2008
Contributed Oral AIP | Quantum Science and Technology (QST)

Description

Quantum machine learning seeks to deliver a decisive computational advantage in data processing by evaluating specific functions of quantum states that remain classically intractable to compute. However, for this advantage to become viable, it is essential to bypass the prohibitively costly characterisation, or full state tomography, of individual data instances. Instead, we must favour the efficient, direct evaluation of their similarities.

In this talk, we present a sample-optimal and hardware-efficient protocol for estimating quantum similarity, specifically the state overlap, using multi-mode bosonic quantum interference. Crucially, the sample complexity of our approach is completely independent of the system’s dimension and is information-theoretically optimal up to a constant factor. This fundamentally departs from the classical intuition that comparing two objects inherently requires characterising them individually first; instead, we exploit quantum mechanics to process joint information directly.

Experimentally, we implement this scheme on Prakash-1, our fully programmable integrated photonic quantum processor. By preparing and interfering qudit states on the chip to directly extract their overlap, we demonstrate the classification and online learning of quantum data with notably high accuracy, even under realistic noisy experimental conditions.

Our findings establish a clear measurement hierarchy, demonstrating that interference-based joint measurements optimally balance ideal sample complexity with practical experimental accessibility. Ultimately, these results establish joint overlap measurements as a scalable pathway to efficient quantum data analysis and serve as a robust, practical building block for network-integrated quantum machine learning. By successfully bypassing the exponentially scaling overheads associated with conventional state tomography methodologies, we showcase how integrated photonic processors can unlock the true potential of quantum machine learning in next-generation networks, bringing these practical quantum advantages significantly closer to reality for real-world data processing applications across various complex scientific domains and advanced engineering challenges we face today, accelerating the transition from theoretical physics to tangible technological tools.

I am the presenting author Yes

Author

Raj Patel (Imperial College London)

Co-authors

Zhenghao Li (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Hao Zhan (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Shana Winston (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Ewan Mer (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Zhenghao Yin (University of Vienna, Faculty of Physics, Vienna Center for Quantum Science and Technology (VCQ), Boltzmanngasse 5, Vienna A-1090, Austria) Shang Yu (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Yazeed Alwehaibi (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Gerard Machado (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Dayne Lopena (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Lijian Zhang (College of Engineering and Applied Sciences, Nanjing University, 163 Xianlin Road, Nanjing 210093, China) Myungshik Kim (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Aonan Zhang (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom) Ian Walmsley (Blackett Laboratory, Department of Physics, Imperial College London, London SW7 2AZ, United Kingdom)

Presentation materials

There are no materials yet.