Authors: Han-Xiao Tao, Jiaqi Hu, Re-Bing Wu

Published on: February 05, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.0288

Summary

• What is new: Investigates pulse-based quantum machine learning models on NISQ devices as an alternative to gate-based models, highlighting their enhanced expressive power.
• Why this is important: Gate-based quantum machine learning models are limited in expressivity due to restrictions in circuit depth within finite coherence time on NISQ devices.
• What the research proposes: Introducing pulse-based models that allow for the construction of ‘infinitely’ deep quantum neural networks within the same coherence time, utilizing quantum control theory.
• Results: Demonstrated that pulse-based models have greater expressive power by approximating arbitrary nonlinear functions and showing improvement in numerical simulations with longer pulse length or more qubits.

Technical Details

Technological frameworks used: Quantum control theory

Models used: Pulse-based quantum neural networks

Data used: Numerical simulations

Potential Impact

This research could influence sectors relying on complex quantum computations like cryptography, pharmaceuticals, and data analytics companies by providing more powerful QML models.

Want to implement this idea in a business?

We have generated a startup concept here: PulseNet Quantum.