QMLP: An Error-Tolerant Nonlinear Quantum MLP Architecture using Parameterized Two-Qubit Gates

TL;DR

This paper introduces QMLP, a quantum neural network architecture that improves accuracy and efficiency on the MNIST dataset by incorporating error-tolerant embedding, enhanced nonlinearity, and parameterized entangling gates, addressing key limitations of current QNNs.

Contribution

QMLP presents a novel quantum neural network design with error-tolerant embedding, richer nonlinearity, and adaptable entanglement, achieving higher accuracy with fewer resources.

Findings

Increases MNIST classification accuracy by 10%

Uses half the quantum gates of prior models

Reduces parameters by threefold

Abstract

Despite potential quantum supremacy, state-of-the-art quantum neural networks (QNNs) suffer from low inference accuracy. First, the current Noisy Intermediate-Scale Quantum (NISQ) devices with high error rates of 0.001 to 0.01 significantly degrade the accuracy of a QNN. Second, although recently proposed Re-Uploading Units (RUUs) introduce some non-linearity into the QNN circuits, the theory behind it is not fully understood. Furthermore, previous RUUs that repeatedly upload original data can only provide marginal accuracy improvements. Third, current QNN circuit ansatz uses fixed two-qubit gates to enforce maximum entanglement capability, making task-specific entanglement tuning impossible, resulting in poor overall performance. In this paper, we propose a Quantum Multilayer Perceptron (QMLP) architecture featured by error-tolerant input embedding, rich nonlinearity, and enhanced variational circuit ansatz with parameterized two-qubit entangling gates. Compared to prior arts, QMLP increases the inference accuracy on the 10-class MNIST dataset by 10% with 2 times fewer quantum gates and 3 times reduced parameters. Our source code is available and can be found in [1]