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

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.

arXiv comment9 pages, 14 images, 6 tables

References in corpus (11)

• Quantum Convolutional Neural Networks
• Parameterized quantum circuits as machine learning models
• Circuit-centric quantum classifiers
• Expressibility and entangling capability of parameterized quantum circuits for hybrid quantum-classical algorithms
• Quantum convolutional neural network for classical data classification
• Hierarchical quantum classifiers
• A divide-and-conquer algorithm for quantum state preparation
• Generalization despite overfitting in quantum machine learning models
• Overfitting in quantum machine learning and entangling dropout
• A preprocessing perspective for quantum machine learning classification advantage using NISQ algorithms
• A General Approach to Dropout in Quantum Neural Networks