Resource Optimal Executable Quantum Circuit Generation Using Approximate Computing

Quantum Computing is an emerging technology that combines the principles of computer science and quantum mechanics to solve computationally challenging problems significantly faster than classical computers. In this paper, we present a proof-of-principle procedure for generating hardware-executable quantum circuits for Noisy Intermediate-Scale Quantum (NISQ) devices that follows the paradigm of approximate computing.Our approach starts from the reference circuit and trans-forms it into an executable circuit with tuneable parameters by replacing the high-level quantum operations by approximate decompositions into hardware-native gates. An inner optimization loop over the rotation gates’ angles ensures that the so-created circuit behaves in the same way as the reference one in terms of its expectation-value landscape. This technique is complemented by compiler-based optimizations to further reduce or aggregate gate groups of the optimized circuit. This three-step procedure is embedded into an outer genetic algorithm framework that inspects many different circuit designs with placements of single- and multi-qubit gates according to the hardware’s lattice structure, and returns a set of approximate quantum circuits that can be executed on NISQ devices directly.We have validated our approach for superconducting quantum systems from IBM and Rigetti for various benchmark algorithms. In nearly all cases, our approach outperforms the vendors’ quantum-compiler frameworks and produces significantly smaller circuits with up to 50% reduction in the number of gates.