ResearchPod Summary
Quantum error mitigation (QEM) is essential for extending the utility of noisy intermediate-scale quantum (NISQ) devices. However, imperfect noise characterization can lead to a sharp breakdown in mitigation performance, where the mitigated signal fails as circuit depth increases. The authors investigate the existence and location of these error-mitigation thresholds across different circuit architectures, specifically focusing on two-dimensional (2D) geometries like square and heavy-hex lattices that are standard in current superconducting quantum processors.
To study these thresholds, the authors map the error-mitigation process onto a statistical-mechanical model, specifically a random-field Ising model (RFIM) in spacetime. They develop a fixed-bond-dimension matrix-product-state (MPS) method to simulate the two-replica transfer dynamics of these circuits. By representing the coefficient vector of the replicated density matrix as an MPS and applying the transfer matrix as a matrix product operator (MPO), they can simulate system sizes and depths significantly beyond the reach of exact numerical propagation. They use the circuit-averaged Renyi-2 entropy and mutual information as diagnostics to locate the phase transition between successful and failed mitigation.
Understanding the stability of error mitigation is critical for the scalability of near-term quantum computing. By establishing a practical tool to probe these thresholds in realistic 2D architectures, this work provides a framework for researchers to predict when error mitigation will fail and to optimize circuit design to remain within the stable regime.
AI-generated third-party summary by ResearchPod. Not official content or an endorsement by the paper authors or affiliated organizations.