Evolutionary-Based Circuit Optimization for Distributed Quantum Computing

Abstract

In this work, we evaluate an evolutionary algorithm (EA) to optimize a given circuit in such a way that it reduces the required communication when executed in the Distributed Quantum Computing (DQC) paradigm. We evaluate our approach for a state preparation task using Grover circuits and show that it is able to reduce the required global gates by more than 89% while still achieving high fidelity as well as the ability to extract the correct solution to the given problem. We also apply the approach to reduce circuit depth and number of CX gates. Additionally, we run experiments in which a circuit is optimized for a given network topology after each qubit has been assigned to specific nodes in the network. In these experiments, the algorithm is able to reduce the communication cost (i.e., number of hops between QPUs) by up to 19%.

Accepted at IEEE QAI 2025

References in corpus (9)

• Towards a Distributed Quantum Computing Ecosystem
• Distributed Quantum Computing: a Survey
• Compiler Design for Distributed Quantum Computing
• Mapping quantum circuits to modular architectures with QUBO
• Optimization of Clifford Circuits
• Generalised Circuit Partitioning for Distributed Quantum Computing
• Reducing the CNOT count for Clifford+T circuits on NISQ architectures
• Evolutionary Quantum Architecture Search for Parametrized Quantum Circuits
• Time-Aware Qubit Assignment and Circuit Optimization for Distributed Quantum Computing