Simultaneous Quantum Machine Learning Training and Architecture Discovery

Abstract

With the onset of gated quantum machine learning, the architecture for such a system is an open question. Many architectures are created either ad hoc or are directly analogous from known classical architectures. Presented here is a novel algorithm which learns a gated quantum machine learning architecture while simultaneously learning its parameters. This proof of concept and some of its variations are explored and discussed.

References in corpus (5)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Synthesis of Quantum Logic Circuits
• Practical optimization for hybrid quantum-classical algorithms
• Gradients of parameterized quantum gates using the parameter-shift rule and gate decomposition
• Decomposition of unitary matrices and quantum gates