A Post-Training Approach for Mitigating Overfitting in Quantum Convolutional Neural Networks

TL;DR

This paper investigates post-training methods to reduce overfitting in quantum convolutional neural networks, highlighting the importance of entanglement and proposing an efficient parameter adaptation technique that improves generalization.

Contribution

It introduces a novel parameter adaptation approach for mitigating overfitting in QCNNs, addressing limitations of classical dropout methods in the quantum context.

Findings

Classical dropout reduces success probability in QCNNs due to entanglement loss.

Proposed parameter adaptation effectively mitigates overfitting.

Entanglement plays a crucial role in QCNN performance.

Abstract

Quantum convolutional neural network (QCNN), an early application for quantum computers in the NISQ era, has been consistently proven successful as a machine learning (ML) algorithm for several tasks with significant accuracy. Derived from its classical counterpart, QCNN is prone to overfitting. Overfitting is a typical shortcoming of ML models that are trained too closely to the availed training dataset and perform relatively poorly on unseen datasets for a similar problem. In this work we study post-training approaches for mitigating overfitting in QCNNs. We find that a straightforward adaptation of a classical post-training method, known as neuron dropout, to the quantum setting leads to a significant and undesirable consequence: a substantial decrease in success probability of the QCNN. We argue that this effect exposes the crucial role of entanglement in QCNNs and the vulnerability of QCNNs to entanglement loss. Hence, we propose a parameter adaptation method as an alternative method. Our method is computationally efficient and is found to successfully handle overfitting in the test cases.