# Experimental demonstration of quantum learning speed-up with classical input data

## TL;DR

This paper demonstrates an optical quantum-classical hybrid machine learning system that achieves a 36% speed-up over classical methods without converting large classical data into quantum states, and shows robustness to noise.

## Contribution

It introduces a quantum learning approach that processes classical input data directly, avoiding data conversion, and experimentally demonstrates a significant speed-up and noise resilience.

## Key findings

- Quantum hybrid machine learning achieves 36% speed-up.
- The system is robust against dephasing noise.
- Experimental validation with optical setup.

## Abstract

We consider quantum-classical hybrid machine learning in which large-scale input channels remain classical and small-scale working channels process quantum operations conditioned on classical input data. This does not require the conversion of classical (big) data to a quantum superposed state, in contrast to recently developed approaches for quantum machine learning. We performed optical experiments to illustrate a single-bit universal machine, which can be extended to a large-bit circuit for binary classification task. Our experimental machine exhibits quantum learning speed-up of approximately 36%, as compared to the fully classical machine. In addition, it features strong robustness against dephasing noise.