Quantum error mitigation improved without costly ideal outputs
Weakly Supervised Quantum Error Mitigation
Artificial Intelligence
Summary
Quantum computers can be very sensitive to errors, and fixing these errors usually requires perfect answers to guide the correction process, which can be very hard to get. This paper shows a way to reduce errors using cheaper, less reliable signals from the quantum device and its setup instead of perfect answers. The approach combines many small indicators to estimate error rates and correct them, working well on real IBM quantum computers. While the method isn't as powerful as those relying on perfect data, it excels where such ideal data cannot be computed.
What this means in practice
- •For quantum hardware teams: Improve noise correction on quantum devices without requiring computationally expensive ideal outputs for training error mitigation models.
- •For quantum algorithm engineers: Reduce output errors in mid-size quantum circuit experiments where classical simulation to get ideal answers is infeasible.
Authors
Seyed Mohamad Ali Tousi, G. N. DeSouza
Abstract
Supervised approaches to quantum error mitigation learn a map from noisy circuit outputs to ideal ones, and therefore require the ideal outputs. Producing those ideal outputs demands noiseless classical simulation, whose cost grows exponentially with system size, so supervision is unavailable in exactly the regime where mitigation matters most. We ask whether cheap, individually unreliable signals drawn from circuit structure and hardware calibration can take the place of ideal labels. We assemble sixteen heuristic labeling functions (stabilizer and parity constraints, relaxation and readout characteristics, local depth, gate counts, and neighboring activity), reconcile their disagreements with a probabilistic label model, and read the resulting per-qubit error probabilities as a readout channel whose inverse mitigates the measured distribution. No ideal output enters the training path. On $147{,}000$ five-qubit circuits executed on two IBM devices, the method removes $24.3\%$ (Algiers) and $28.8\%$ (Hanoi) of the Kullback-Leibler divergence to the ideal distribution, against $15.4\%$ and $21.5\%$ for the strongest published analytical baseline, a margin that holds on both devices and lies far outside its bootstrap interval. Supervised neural models trained on ideal distributions remain stronger where such labels exist, and we quantify that gap rather than setting it aside; the method's claim is to the regime where they do not, since the labels they require cannot be computed for the circuits mitigation is needed for. The codes will be released shortly.