Grid-Partitioned MWIS Solving with Neutral Atom Quantum Computing for QUBO Problems

TL;DR

This paper introduces a hybrid quantum-classical method using neutral atom quantum computing and spatial grid partitioning to efficiently approximate solutions to large-scale QUBO problems, demonstrated on portfolio optimization.

Contribution

It presents a novel framework combining neutral atom quantum computing with grid partitioning to solve large QUBO problems more efficiently than classical methods.

Findings

Competitive performance against simulated annealing.

Scalability demonstrated on 50-asset portfolio optimization.

Potential for practical large-scale optimization in NISQ era.

Abstract

Quadratic Unconstrained Binary Optimization (QUBO) problems are prevalent in real-world applications, such as portfolio optimization, but pose significant computational challenges for large-scale instances. We propose a hybrid quantum-classical framework that leverages neutral atom quantum computing to address QUBO problems by mapping them to the Maximum Weighted Independent Set (MWIS) problem on unit disk graphs. Our approach employs spatial grid partitioning to decompose the problem into manageable subgraphs, solves each subgraph using Analog Hamiltonian Simulation (AHS), and merges solutions greedily to approximate the global optimum. We evaluate the framework on a 50-asset portfolio optimization problem using historical S&P 500 data, benchmarking against classical simulated annealing. Results demonstrate competitive performance, highlighting the scalability and practical potential of our method in the Noisy Intermediate-Scale Quantum (NISQ) era. As neutral atom quantum hardware advances, our framework offers a promising path toward solving large-scale optimization problems efficiently.