A Multiobjective Approach for Nearest Neighbor Optimization of N-Dimensional Quantum Circuits

Benjamín Barán, Alejo Carballude, Marcos Villagra

Abstract: The Nearest Neighbor (NN) restriction in quantum circuits requires quantum gates to act on geometrically adjacent qubits. Methods that convert generic quantum circuits and allow them to comply with the NN restriction have already been studied in the literature, where the main technique to accomplishing this task is by inserting SWAP gates into the circuit. In previous works, other authors have introduced a two-dimensional multi-objective NN conversion algorithm that takes into account two simultaneous objectives: the minimization of the two-dimensional grid size and the minimization of the number of SWAP gates required to allow any generic circuit to comply with the NN restriction. An extended higher-dimensional version of these previous methods is presented in this work, maintaining the same optimization objectives. We present experimental results for three-dimensional circuits, which show improvements for 52.6% of the tested circuits over all previously published results to date.

External IDs:dblp:journals/sncs/BaranCV21