54-Qubit Device Validates Noise-Resilient Optimization Framework

A 54-qubit IQM Emerald quantum device has validated a new optimization framework that achieves competitive solutions using a surprisingly shallow circuit depth. Researchers Elisabeth Wybo and a colleague developed Quantum-Informed Surrogate Sampling, or QISS, which leverages a quantum computer to generate “informative statistics for scalable classical sampling” rather than directly solving problems. The work demonstrates that QISS, using only O(N) low-order correlators from shallow circuits, can outperform the widely studied QAOA algorithm; specifically, on MaxCut problems with 3-regular graphs, QISS from p=3 QAOA correlators outperforms vanilla QAOA at p=17 on average. This approach, validated through experiments, suggests a path toward noise-resilient near-term quantum optimization.

This performance is notable because a shallow circuit’s capacity can exceed a much deeper one through effective post-processing. The framework does not directly sample solutions using the quantum computer, but instead generates “informative statistics for scalable classical sampling,” a shift that may prove crucial for near-term quantum optimization. Validation on the IQM Emerald device further confirms QISS’s noise resilience, suggesting a viable path forward for practical quantum optimization strategies. The researchers report that further improvements are possible by using QISS to warm-start QAOA, potentially unlocking even greater performance gains.

Researchers are shifting strategies in the pursuit of near-term quantum optimization, focusing on leveraging shallow circuits to enhance classical algorithms instead of relying on direct quantum sampling. The efficiency of QISS stems from its reliance on only O(N) low-order correlators, allowing it to achieve competitive results on problems like MaxCut and Maximum Independent Set. This approach, detailed in their recent paper, supports a model where shallow quantum circuits generate data for classical processing, offering a viable path toward scalable and noise-resilient optimization.