Quantum Error Propagation

TL;DR

This paper investigates whether quantum machine learning models are more resistant to data poisoning attacks due to the unique properties of quantum information processing, such as qubit confinement and unitary operations.

Contribution

It hypothesizes and tests that error propagation in quantum models is limited, contrasting with classical models, due to quantum-specific features.

Findings

Error propagation in quantum models appears limited.

Quantum encoding confines data to the Bloch sphere.

Unitary operations preserve quantum information integrity.

Abstract

Data poisoning attacks on machine learning models aim to manipulate the data used for model training such that the trained model behaves in the attacker's favour. In classical models such as deep neural networks, large chains of dot products do indeed cause errors injected by an attacker to propagate or accumulate. But what about quantum models? We hypothesise that, in quantum machine learning, error propagation is limited for two reasons. The first is that data, which is encoded in quantum computing, is in terms of qubits that are confined to the Bloch sphere. Second, quantum information processing happens via the application of unitary operators, which preserve norms. Testing this hypothesis, we investigate how extensive error propagation and, thus, poisoning attacks affect quantum machine learning.