Saved in:
Bibliographic Details
Main Authors: Punch, Samuel, Guha, Krishnendu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.00811
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912562757500928
author Punch, Samuel
Guha, Krishnendu
author_facet Punch, Samuel
Guha, Krishnendu
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2509_00811
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MAESTROCUT: Dynamic, Noise-Adaptive, and Secure Quantum Circuit Cutting on Near-Term Hardware
Punch, Samuel
Guha, Krishnendu
Cryptography and Security
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.
title MAESTROCUT: Dynamic, Noise-Adaptive, and Secure Quantum Circuit Cutting on Near-Term Hardware
topic Cryptography and Security
url https://arxiv.org/abs/2509.00811