Protocol for secure quantum machine learning at a distant place

TL;DR

This paper proposes a secure quantum machine learning protocol enabling a remote provider to teach an arbitrarily initialized device, with built-in security features to detect or prevent external interference, demonstrated on single-qubit devices.

Contribution

It introduces a novel remote quantum learning protocol with security measures, advancing secure quantum machine learning methods.

Findings

Protocol successfully teaches single-qubit devices

Security features detect external interference

Trade-off between accuracy and learning time

Abstract

The application of machine learning to quantum information processing has recently attracted keen interest, particularly for the optimization of control parameters in quantum tasks without any pre-programmed knowledge. By adapting the machine learning technique, we present a novel protocol in which an arbitrarily initialized device at a learner's location is taught by a provider located at a distant place. The protocol is designed such that any external learner who attempts to participate in or disrupt the learning process can be prohibited or noticed. We numerically demonstrate that our protocol works faithfully for single-qubit operation devices. A trade-off between the inaccuracy and the learning time is also analyzed.