Quantum Machine Learning for Industrial Applications
A new thesis explores Quantum Machine Learning (QML) for industrial applications, addressing challenges in trainability, expressivity, and classical simulation resistance. It introduces subspace-preserving QML algorithms, including photonic circuits and quantum convolutional neural networks, designed to offer polynomial quantum advantage. The research also analyzes variational quantum circuits as quantum Fourier models, establishing conditions for quantum models to provably separate from classical counterparts. AI