Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Resource-Observability Dual-Budget Control for QEC under Resource-Constrained Stages

Duy Hoang Nguyen

Published Mar 16, 2026• 0 citations
Publisher

Abstract

In some embodiments, disclosed systems and methods provide resource-aware scheduling of quantum error-correction (QEC) stabilizer measurements at a resource-constrained stage. A classical controller obtains stage state data indicative of resource availability and, for each control interval, determines a measurement execution policy constrained by the stage state data (in some embodiments, comprising a measurement schedule and/or per-stabilizer execution rates or probabilities). The controller computes one or more observability-risk indicators that quantify accumulated loss of observability (or a quantitative proxy thereof), mismatch between a predictive model and observed syndrome data, and/or uncertainty of a decoder or estimator output. An observability guard enforces bounded observability by causing a full-observation operation when elapsed time since a most recent full-observation operation reaches a threshold T_max and/or when drift or elevated risk is detected, and may do so even when a controller-layer resource guard would otherwise restrict execution due to reduced resource availability, while remaining subject to an independent physical safety interlock or absolute cutoff external to the controller, when provided. In some embodiments, the resource constraint is implemented using a resource model and a time-varying resource headroom value R(t) with predicted constrained-resource consumption constrained by R(t). In some embodiments, the resource constraint is implemented using a resource-availability indicator derived directly from stage state data and reactive adaptation without requiring prediction. In some embodiments, the resource-availability indicator is derived from a computed headroom value R(t), for example as a thresholded or quantized headroom class of R(t). In some embodiments, a dual-budget formulation tracks both resource-budget state and observability-risk budget state (e.g., resource tokens and QEC-risk tokens), optionally using a sentinel subset to maintain drift visibility and to update an error/noise model under constrained resources. In a non-limiting cryogenic embodiment, stage state data comprises cryogenic state data, R(t)=H(t) comprises cooling headroom, and resource-budget state may be represented using thermal tokens.In some embodiments, the observability-risk indicator comprises a model uncertainty measure.In some embodiments, the observability guard initiates the full-observation operation according to a predetermined or randomized refresh schedule that is independent of the stage state data and independent of the resource-availability indicator, while still being subject to any independent physical safety interlock or absolute cutoff external to the classical controller, when provided.In some embodiments, the observability guard initiates the full-observation operation based on a stochastic refresh mechanism in which inter-refresh intervals are drawn from a configured probability distribution and/or are generated by a pseudorandom process, without requiring a comparison of elapsed time to a fixed threshold. In some cases, the configured probability distribution has at least one parameter that is configurable based on a drift metric and/or the observability-risk indicator.In some embodiments, the system optionally includes an independent physical safety interlock or absolute cutoff external to the classical controller; when provided, initiation and/or execution of the full-observation operation is conditioned on the interlock or cutoff, and when not provided the observability guard may initiate the full-observation operation without such conditioning.In some embodiments, non-limiting method overview (example steps): (i) obtain stage state data; (ii) derive a resource constraint from the stage state data, including by computing resource headroom R(t) using a resource model or by deriving a resource-availability indicator without using a resource model; (iii) determine or update a measurement execution policy for a control interval subject to the resource constraint; (iv) compute one or more observability-risk indicators; (v) evaluate observability-guard conditions, including a time-based threshold T_max and/or drift/risk criteria; (vi) execute stabilizer measurements according to the measurement execution policy, optionally including a sentinel subset; (vii) when triggered, execute a full-observation operation, including as a multi-interval refresh sequence; (viii) compute deviation and/or drift metrics; (ix) update an error/noise model and optionally resource-model parameters and/or budget state; and (x) repeat for subsequent control intervals.

DOI
10.5281/zenodo.19041531

Related papers

Method and System for Gated Identifiability Validation of Phase Offset and/or Phase Lag in a Quantum System Prior to Parameter Update

Duy Hoang Nguyen

The invention relates to calibration, validation, and control of a quantum system, such as a quantum processor, a quantum sensor, or a quantum measurement apparatus. More particularly, the invention provides a computer-implemented method and system for determining whether a phase-offset and/or phase-lag indicator derived from at least two observation streams is sufficiently identifiable for use in inference and/or parameter update. Rather than permitting direct use of the indicator once computed, the invention requires passage of a machine-enforceable identifiability gate Iφ before the indicator is admitted for downstream use. In preferred implementations, the method defines at least a first measurement role and a second measurement role, acquires observation data under the two roles according to a role-activation schedule that prevents simultaneous access within the same acquisition step, and computes at least one phase-offset/phase-lag indicator over a candidate lag window. The identifiability gate Iφ then applies multiple technical checks, including calibration or estimation of acquisition-chain latency, instrument swap and/or acquisition-chain swap to test invariance of the indicator, at least one negative control, and explicit failure criteria for declaring the indicator not identifiable. Only when the indicator passes the gate is it used to generate a validated output and to update at least one control, acquisition, or signal pre-processing parameter of the quantum system. Upon failure of the gate, any parameter update based on the indicator is blocked and diagnostic data are output. In more specific embodiments, role switching is accepted only when predefined micro-invariants Q are preserved within tolerance, and a Van-proxy role-switch action is selected only if it passes a Van-proxy audit configured to reject actions that perturb relevant micro-dynamics of the quantum system. Additional embodiments may employ an adaptive role-activation schedule, a macro-entropy-based drift or instability signal, a frozen configuration manifest with version-specific hash binding, and a parameter-update lock that remains active until a new identifiability cycle passes after drift or failure. The invention thereby reduces artifact-driven parameter updates, improves robustness against acquisition-chain asymmetry and processing artifacts, and provides controlled requalification before re-enabling update.

•Mar 16, 2026•0 citations

SAGE-QEC: Structure-Aware Gated Error Correction for Surface-Code Memories Under Structured Noise

Asia Alhammadi, Mohammed Al-Absi

Abstract Quantum error-correction decoders must convert noisy syndrome records into logical decisions while operating under strict latency and reliability constraints. Minimum-weight perfect matching (MWPM) remains a strong baseline for surface-code decoding because it is fast, interpretable, and exact for its graphlike optimization problem [1–8]. However, the graphlike independent-error approximation can become mismatched when syndrome records contain temporal drift, measurement corruption, crosstalk, bursts, or mixtures of mechanisms [9–18]. We introduce SAGE-QEC, a failure-aware selective decoder that retains nominal PyMatching as its default decision and learns when an observable syndrome structure warrants an alternative candidate. SAGE-QEC uses train-frozen structure thresholds, a candidate ranker, validation-only policy selection, and a paired blind evaluation. Across 45 simulator conditions, SAGE-QEC reduced the aggregate logical error rate from 27.1285% for PyMatching to 24.8320%, corresponding to an 8.47% relative reduction. The campaign used rotated surface-code memory circuits at distances 3, 5, and 7, nominal circuit parameters 0.002, 0.003, and 0.005, five seeds, and four test regimes withheld from training: Burst-3, Temporal Drift, Crosstalk, and Mixed Noise. The test set contained 81,000 paired shots. All twelve distance-by-regime aggregate comparisons favored SAGE-QEC in the executed campaign, with paired bootstrap intervals below zero for SAGE-QEC minus PyMatching. The result is evidence of a reproducible simulator-level improvement, not a claim of universal decoder superiority. SAGE-QEC currently uses logical-output proxy candidates rather than a complete physical correction search, and its intervention rate remains high under distribution shift. It has not yet been compared with a fully calibrated correlated-noise space-time PyMatching baseline or hardware calibration data. We therefore characterize SAGE-QEC as a strong qualification-stage prototype and release a complete audit protocol for the next industrial validation stage.

•Sep 9, 2026•0 citations

System and method of optimizing instructions for quantum computers

Frederic T. Chong, I. Yunong Shi, Nelson Leung +3 more

A quantum computing system includes a quantum processor having a plurality of qubits, a classical memory, and a classical processor. The classical processor is configured to compile a quantum program into logical assembly instructions in an intermediate language, aggregate the logical assembly instructions together into a plurality of logical blocks of instructions, generate a logical schedule for the quantum program based on commutativity between the plurality of logical blocks, generate a tentative physical schedule based on the logical schedule, the tentative physical schedule includes a mapping of the logical assembly instructions in the logical schedule onto the plurality of qubits of the quantum processor, aggregate instructions together within the tentative physical schedule that do not reduce parallelism, thereby generating an updated physical schedule; generate optimized control pulses for the aggregated instructions, and execute the quantum program on the quantum processor with the optimized control pulses and the updated physical schedule.

•Sep 12, 2019•0 citations

SYSTEM AND METHOD OF OPTIMIZING INSTRUCTIONS FOR QUANTUM COMPUTERS

Frederic T. Chong, I. Yunong Shi, Nelson Leung +3 more

A quantum computing system includes a quantum processor having a plurality of qubits, a classical memory, and a classical processor. The classical processor is configured to compile a quantum program into logical assembly instructions in an intermediate language, aggregate the logical assembly instructions together into a plurality of logical blocks of instructions, generate a logical schedule for the quantum program based on commutativity between the plurality of logical blocks, generate a tentative physical schedule based on the logical schedule, the tentative physical schedule includes a mapping of the logical assembly instructions in the logical schedule onto the plurality of qubits of the quantum processor, aggregate instructions together within the tentative physical schedule that do not reduce parallelism, thereby generating an updated physical schedule; generate optimized control pulses for the aggregated instructions, and execute the quantum program on the quantum processor with the optimized control pulses and the updated physical schedule.

•Sep 12, 2019•0 citations

Reinforcement Learning-Based Adaptive Quantum-Safe Cryptography for DN25-Compliant Smart Environments

Darlan Noetzold, Jorge L. V. Barbosa, Juan F. De Paz +1 more

The emergence of quantum computing challenges traditional security mechanisms, particularly in heterogeneous and resource-constrained IoT and smart environments that must satisfy DN25 requirements. This work introduces a reinforcement learning-driven model for the adaptive selection and orchestration of cryptographic algorithms. Acting as an intelligent decision layer, the system observes contextual, network, and operational metrics to recommend or enforce configurations combining classical schemes, post-quantum cryptography, and Quantum Key Distribution when available. The selection problem is formulated as a Markov Decision Process with state and action spaces aligned with protocol control flows and is embedded into a security middleware with negotiation and fallback mechanisms to ensure interoperability and policy compliance without modifying application logic. Experimental results demonstrate that the model dynamically adjusts key lengths, algorithm families, and security policies according to risk and resource conditions, increasing post-quantum cryptography and Quantum Key Distribution usage by up to 33.4% and 23.9% in high-risk scenarios while favoring low-latency classical or hybrid options for less critical traffic. The system achieves success rates above 78% while maintaining stable latency and resource usage during extended operation.

IEEE Access•Jan 1, 2026•1 citation

Temporal Susceptibility Feedback for Hardware-efficient Bosonic Quantum Error Correction beyond Markovian Approximation

Isamu Ohnishi

•Jan 1, 2026•0 citations
In some embodiments, disclosed systems and methods provide resource-aware scheduling of quantum error-correction (QEC) stabilizer measurements at a resource-constrained stage. A classical controller obtains stage state data indicative of resource availability and, for each control interval, determines a measurement execution policy constrained by the stage state data (in some embodiments, comprising a measurement schedule and/or per-stabilizer execution rates or probabilities). The controller computes one or more observability-risk indicators that quantify accumulated loss of observability (or a quantitative proxy thereof), mismatch between a predictive model and observed syndrome data, and/or uncertainty of a decoder or estimator output. An observability guard enforces bounded observability by causing a full-observation operation when elapsed time since a most recent full-observation operation reaches a threshold T_max and/or when drift or elevated risk is detected, and may do so even when a controller-layer resource guard would otherwise restrict execution due to reduced resource availability, while remaining subject to an independent physical safety interlock or absolute cutoff external to the controller, when provided. In some embodiments, the resource constraint is implemented using a resource model and a time-varying resource headroom value R(t) with predicted constrained-resource consumption constrained by R(t). In some embodiments, the resource constraint is implemented using a resource-availability indicator derived directly from stage state data and reactive adaptation without requiring prediction. In some embodiments, the resource-availability indicator is derived from a computed headroom value R(t), for example as a thresholded or quantized headroom class of R(t). In some embodiments, a dual-budget formulation tracks both resource-budget state and observability-risk budget state (e.g., resource tokens and QEC-risk tokens), optionally using a sentinel subset to maintain drift visibility and to update an error/noise model under constrained resources. In a non-limiting cryogenic embodiment, stage state data comprises cryogenic state data, R(t)=H(t) comprises cooling headroom, and resource-budget state may be represented using thermal tokens.In some embodiments, the observability-risk indicator comprises a model uncertainty measure.In some embodiments, the observability guard initiates the full-observation operation according to a predetermined or randomized refresh schedule that is independent of the stage state data and independent of the resource-availability indicator, while still being subject to any independent physical safety interlock or absolute cutoff external to the classical controller, when provided.In some embodiments, the observability guard initiates the full-observation operation based on a stochastic refresh mechanism in which inter-refresh intervals are drawn from a configured probability distribution and/or are generated by a pseudorandom process, without requiring a comparison of elapsed time to a fixed threshold. In some cases, the configured probability distribution has at least one parameter that is configurable based on a drift metric and/or the observability-risk indicator.In some embodiments, the system optionally includes an independent physical safety interlock or absolute cutoff external to the classical controller; when provided, initiation and/or execution of the full-observation operation is conditioned on the interlock or cutoff, and when not provided the observability guard may initiate the full-observation operation without such conditioning.In some embodiments, non-limiting method overview (example steps): (i) obtain stage state data; (ii) derive a resource constraint from the stage state data, including by computing resource headroom R(t) using a resource model or by deriving a resource-availability indicator without using a resource model; (iii) determine or update a measurement execution policy for a control interval subject to the resource constraint; (iv) compute one or more observability-risk indicators; (v) evaluate observability-guard conditions, including a time-based threshold T_max and/or drift/risk criteria; (vi) execute stabilizer measurements according to the measurement execution policy, optionally including a sentinel subset; (vii) when triggered, execute a full-observation operation, including as a multi-interval refresh sequence; (viii) compute deviation and/or drift metrics; (ix) update an error/noise model and optionally resource-model parameters and/or budget state; and (x) repeat for subsequent control intervals.
The invention relates to calibration, validation, and control of a quantum system, such as a quantum processor, a quantum sensor, or a quantum measurement apparatus. More particularly, the invention provides a computer-implemented method and system for determining whether a phase-offset and/or phase-lag indicator derived from at least two observation streams is sufficiently identifiable for use in inference and/or parameter update. Rather than permitting direct use of the indicator once computed, the invention requires passage of a machine-enforceable identifiability gate Iφ before the indicator is admitted for downstream use. In preferred implementations, the method defines at least a first measurement role and a second measurement role, acquires observation data under the two roles according to a role-activation schedule that prevents simultaneous access within the same acquisition step, and computes at least one phase-offset/phase-lag indicator over a candidate lag window. The identifiability gate Iφ then applies multiple technical checks, including calibration or estimation of acquisition-chain latency, instrument swap and/or acquisition-chain swap to test invariance of the indicator, at least one negative control, and explicit failure criteria for declaring the indicator not identifiable. Only when the indicator passes the gate is it used to generate a validated output and to update at least one control, acquisition, or signal pre-processing parameter of the quantum system. Upon failure of the gate, any parameter update based on the indicator is blocked and diagnostic data are output. In more specific embodiments, role switching is accepted only when predefined micro-invariants Q are preserved within tolerance, and a Van-proxy role-switch action is selected only if it passes a Van-proxy audit configured to reject actions that perturb relevant micro-dynamics of the quantum system. Additional embodiments may employ an adaptive role-activation schedule, a macro-entropy-based drift or instability signal, a frozen configuration manifest with version-specific hash binding, and a parameter-update lock that remains active until a new identifiability cycle passes after drift or failure. The invention thereby reduces artifact-driven parameter updates, improves robustness against acquisition-chain asymmetry and processing artifacts, and provides controlled requalification before re-enabling update.
Abstract Quantum error-correction decoders must convert noisy syndrome records into logical decisions while operating under strict latency and reliability constraints. Minimum-weight perfect matching (MWPM) remains a strong baseline for surface-code decoding because it is fast, interpretable, and exact for its graphlike optimization problem [1–8]. However, the graphlike independent-error approximation can become mismatched when syndrome records contain temporal drift, measurement corruption, crosstalk, bursts, or mixtures of mechanisms [9–18]. We introduce SAGE-QEC, a failure-aware selective decoder that retains nominal PyMatching as its default decision and learns when an observable syndrome structure warrants an alternative candidate. SAGE-QEC uses train-frozen structure thresholds, a candidate ranker, validation-only policy selection, and a paired blind evaluation. Across 45 simulator conditions, SAGE-QEC reduced the aggregate logical error rate from 27.1285% for PyMatching to 24.8320%, corresponding to an 8.47% relative reduction. The campaign used rotated surface-code memory circuits at distances 3, 5, and 7, nominal circuit parameters 0.002, 0.003, and 0.005, five seeds, and four test regimes withheld from training: Burst-3, Temporal Drift, Crosstalk, and Mixed Noise. The test set contained 81,000 paired shots. All twelve distance-by-regime aggregate comparisons favored SAGE-QEC in the executed campaign, with paired bootstrap intervals below zero for SAGE-QEC minus PyMatching. The result is evidence of a reproducible simulator-level improvement, not a claim of universal decoder superiority. SAGE-QEC currently uses logical-output proxy candidates rather than a complete physical correction search, and its intervention rate remains high under distribution shift. It has not yet been compared with a fully calibrated correlated-noise space-time PyMatching baseline or hardware calibration data. We therefore characterize SAGE-QEC as a strong qualification-stage prototype and release a complete audit protocol for the next industrial validation stage.
A quantum computing system includes a quantum processor having a plurality of qubits, a classical memory, and a classical processor. The classical processor is configured to compile a quantum program into logical assembly instructions in an intermediate language, aggregate the logical assembly instructions together into a plurality of logical blocks of instructions, generate a logical schedule for the quantum program based on commutativity between the plurality of logical blocks, generate a tentative physical schedule based on the logical schedule, the tentative physical schedule includes a mapping of the logical assembly instructions in the logical schedule onto the plurality of qubits of the quantum processor, aggregate instructions together within the tentative physical schedule that do not reduce parallelism, thereby generating an updated physical schedule; generate optimized control pulses for the aggregated instructions, and execute the quantum program on the quantum processor with the optimized control pulses and the updated physical schedule.
A quantum computing system includes a quantum processor having a plurality of qubits, a classical memory, and a classical processor. The classical processor is configured to compile a quantum program into logical assembly instructions in an intermediate language, aggregate the logical assembly instructions together into a plurality of logical blocks of instructions, generate a logical schedule for the quantum program based on commutativity between the plurality of logical blocks, generate a tentative physical schedule based on the logical schedule, the tentative physical schedule includes a mapping of the logical assembly instructions in the logical schedule onto the plurality of qubits of the quantum processor, aggregate instructions together within the tentative physical schedule that do not reduce parallelism, thereby generating an updated physical schedule; generate optimized control pulses for the aggregated instructions, and execute the quantum program on the quantum processor with the optimized control pulses and the updated physical schedule.
The emergence of quantum computing challenges traditional security mechanisms, particularly in heterogeneous and resource-constrained IoT and smart environments that must satisfy DN25 requirements. This work introduces a reinforcement learning-driven model for the adaptive selection and orchestration of cryptographic algorithms. Acting as an intelligent decision layer, the system observes contextual, network, and operational metrics to recommend or enforce configurations combining classical schemes, post-quantum cryptography, and Quantum Key Distribution when available. The selection problem is formulated as a Markov Decision Process with state and action spaces aligned with protocol control flows and is embedded into a security middleware with negotiation and fallback mechanisms to ensure interoperability and policy compliance without modifying application logic. Experimental results demonstrate that the model dynamically adjusts key lengths, algorithm families, and security policies according to risk and resource conditions, increasing post-quantum cryptography and Quantum Key Distribution usage by up to 33.4% and 23.9% in high-risk scenarios while favoring low-latency classical or hybrid options for less critical traffic. The system achieves success rates above 78% while maintaining stable latency and resource usage during extended operation.