Graduate Defense: Jack Cunningham

Dissertation Information

Title: Combining Quantum Convolutional Neural Networks with Parameter Initialization Techniques to Enhance Barren Plateau
Mitigation

Program: Master of Science in Computer Science

Advisor: Dr. Jun Zhuang, Materials Science and Engineering

Committee Members: Dr. Anne Egger, Geosciences; Dr. Edoaro Serra, Computer Science and Dr. Grady Wright, Mathmatics

Abstract

“In recent years, Variational Quantum Circuit (VQC)s have been widely explored to advance quantum circuits against classic models on various domains, such as quantum chemistry and quantum machine learning. Similar to classic machine-learning models, VQCs can be optimized through gradient-based approaches. However, the gradient of VQCs may dramatically vanish during the optimization process as the number of qubits or layers increases. This issue, a.k.a. Barren Plateau (BP)s, seriously hinders the scaling of VQCs on large datasets. To mitigate the exponential gradient vanishing, extensive efforts have been devoted to tackling this issue through diverse strategies. Some strategies include VQC initialization and new VQC architectures such as Quantum Neural Network (QNN)s.

In this thesis, we aim to combine multiple VQC parameter initialization strategies with the Quantum Convolutional Neural Network (QCNN) architecture. We study the performance of circuit architectures in combination with various initialization techniques. In most existing studies, researchers typically focus on improving or understanding one aspect of the VQC approach. Instead, we plan to investigate how the two components of the VQC interact with each other as we increase the complexities. This investigation will provide a better understanding of how different circuit architectures are affected by initialization strategies and vice versa”