Quantum evolution kernel: Machine learning on graphs with programmable arrays of qubits

The rapid development of reliable quantum processing units opens up novel computational opportunities for machine learning. Here, we introduce a procedure for measuring the similarity between graph-structured data, based on the time evolution of a quantum system. By encoding the topology of the input graph in the Hamiltonian of the system, the evolution produces measurement samples that retain key features of the data. We study analytically the procedure and illustrate its versatility in providing links to standard classical approaches. We then show numerically that this scheme performs well compared to standard graph kernels on typical benchmark datasets. Finally, we study the possibility of a concrete implementation on a realistic neutral-atom quantum processor.

Automated summary

The rapid development of reliable quantum processing units has opened up novel computational opportunities for machine learning. This study introduces a procedure for measuring the similarity between graph-structured data based on the time evolution of a quantum system, which shows promising performance compared to standard graph kernels on benchmark datasets. The possibility of implementing this procedure on a realistic neutral-atom quantum processor is also studied.