Score: 0

MAESTROCUT: Dynamic, Noise-Adaptive, and Secure Quantum Circuit Cutting on Near-Term Hardware

Published: August 31, 2025 | arXiv ID: 2509.00811v1

By: Samuel Punch, Krishnendu Guha

Potential Business Impact:

Makes quantum computers work better and faster.

Business Areas:
Quantum Computing Science and Engineering

We present MaestroCut, a closed-loop framework for quantum circuit cutting that adapts partitioning and shot allocation to device drift and workload variation. MaestroCut tracks a variance proxy in real time, triggers re-cutting when accuracy degrades, and routes shots using topology-aware priors. An online estimator cascade (MLE, Bayesian, GP-assisted) selects the lowest-error reconstruction within a fixed budget. Tier-1 simulations show consistent variance contraction and reduced mean-squared error versus uniform and proportional baselines. Tier-2 emulation with realistic queueing and noise demonstrates stable latency targets, high reliability, and ~1% software overhead under stress scenarios. These results indicate that adaptive circuit cutting can provide accuracy and efficiency improvements with minimal operational cost on near-term hardware.

Country of Origin
🇮🇪 Ireland

Page Count
14 pages

Category
Computer Science:
Cryptography and Security