A clustering aggregation algorithm on neutral-atoms and annealing quantum processors

This work presents a hybrid quantum-classical algorithm to perform clustering aggregation, designed for neutral-atoms and quantum annealing processors. Clustering aggregation is a technique that mitigates the weaknesses of clustering algorithms, an important class of data science methods to partition datasets, widely employed in many industries. By expressing the problem as a Maximum Independent Set (MIS) problem and as a Quadratic Unconstrained Binary Optimization (QUBO) problem with additional constraints, it was possible to submit it to Pasqal's Fresnel (neutral-atoms processor) and D-Wave's Advantage QPU (quantum annealer); additionally, the algorithm was also tested on an emulation of Fresnel running on the QuTiP framework. Results revealed technical limitations, such as the difficulty of adding additional constraints on the employed neutral-atoms platform, the necessity of better metrics to measure the quality of produced clusterings, and the slow performance on the emulator. However, findings suggest a promising potential for future advancements in hybrid quantum-classical pipelines, though further improvements are needed in both quantum and classical components.