Evaluating Neural Network Decoder Performance for Quantum Error Correction Using Various Data Generation Models

Neural networks have been shown in the past to perform quantum error correction (QEC) decoding with greater accuracy and efficiency than algorithmic decoders. Because the qubits in a quantum computer are volatile and only usable on the order of milliseconds before they decohere, a means of fast quantum error correction is necessary in order to correct data qubit errors within the time budget of a quantum algorithm. Algorithmic decoders are good at resolving errors on logical qubits with only a few data qubits, but are less efficient in systems containing more data qubits. With neural network decoders, practical quantum computation becomes much more realizable since the error corrective operations are calculated much faster than with the MWPM or partial lookup table implementations. This research is aimed at furthering neural network QEC decoder research by generating exhaustive and randomly sampled data sets using high-performance computing algorithms to evaluate the effect of data set generation methods on the effectiveness of these neural networks compared to similar models. The results of this work show that different data sets affect various performance metrics including accuracy, F1 score, area under the receiver operating characteristic curve, and QEC cycles