Carnegie Mellon University, United States of America
Optimization of quantum circuits is an integral part of the quantum computing toolchain. In many Noisy Intermediate-Scale Quantum (NISQ) devices, only loose connectivity between qubits is maintained, meaning a valid quantum circuit often requires swapping physical qubits in order to satisfy adjacency requirements. Optimizing circuits to minimize such swaps, as well as other parameters, is imperative for utilizing the quantum hardware of the near future. In this work, we leverage SPIRAL, a code generation system for linear transforms built on GAP’s computer algebra system, and present an application towards optimizing quantum circuits. SPIRAL natively understands tensor products, complex matrices and symbolic matrices, and its proven decomposition and rewriting capabilities are uniquely predisposed to optimize quantum circuits. Specifically, we construct a search problem that can be solved through dynamic programming. The optimal circuit can then be translated to QASM code, where it is executed on a real quantum device.