Hierarchical quantum circuit representations for neural architecture search

Abstract Quantum circuit algorithms often require architectural design choices analogous to those made in constructing neural and tensor networks. These tend be hierarchical, modular exhibit repeating patterns. Neural Architecture Search (NAS) attempts automate network through learning architecture achieves state-of-the-art performance. We propose a framework for representing quantum architectures using techniques from NAS, which enables search space search. use this justify the importance of machine by generating family Convolutional Networks (QCNNs) evaluating them on music genre classification dataset, GTZAN. Furthermore, we employ genetic algorithm perform Phase Recognition (QPR) as an example with our representation. Finally, implement open-source Python package enable dynamic creation facilitate NAS.