arXiv subjects

Yeong-Luh Ueng

Publications and source records attributed to Yeong-Luh Ueng.

3 recordsLinked to original sources

arXiv2026

Lottery BP: Unlocking Quantum Error Decoding at Scale

During a QEC cycle, quantum error decoding stands on the critical path. To enable fault tolerance on millions of qubits in real time, scalable decoding is necessary, which motivates this paper. Existing decoding algorithms (decoders), such as clustering, matching, belief propagation (BP), and neural networks, suffer from one or more of inaccuracy, costliness, and incompatibility, upon a broad set of quantum error correction codes, such as surface code and bivariate bicycle code. Therefore, there exists a gap between existing decoders and an ideal decoder that is accurate, fast, general, and scalable simultaneously. To move closer to the goal above, this paper contributes in three aspects, including decoder algorithm, decoder architecture, and decoding simulator. First, we propose Lottery BP, a lightweight decoder that introduces guided randomness to break the symmetric deadlock caused by quantum degeneracy during decoding. Lottery BP improves the decoding accuracy over BP by up to 6 orders. Second, we design a PolyQec architecture that implements Lottery BP as a local decoder and ordered statistics decoding (OSD) as a global decoder, exemplifying a hierarchical decoder architecture. PolyQec is configurable for surface code and X/Z check. Since Lottery BP boosts the local decoding accuracy, PolyQec invokes the costly global OSD decoder less frequently over BP+OSD to enhance the scalability, e.g., up to 4 orders of magnitude less for surface codes. Third, we develop Syndrilla, a modular PyTorch-based decoding simulator that enables fair, extensible decoder evaluation with unified accuracy and performance metrics. On GPUs, Syndrilla runs 1 order of magnitude faster than CUDAQX.

cs.AR↗

arXiv2026

Deep-Learning-Aided Successive Cancellation List Flip Decoding for Polar Codes

Polar codes are the first error-correcting code proven to achieve channel capacity based on infinite code length. The Successive Cancellation List Flip (SCLF) decoding algorithm was proposed by flipping an erroneous bit during the next decoding attempt. To identify the erroneous bits, the Log-Likelihood Ratio (LLR) is used to indicate the reliability of each decision bit. To improve the accuracy of the erroneous bit prediction, we propose deep-learning-aided (DL-aided) SCLF decoding algorithms. We first offer a stacked LSTM network that contains new features to train our models, which are able to improve the accuracy of the prediction of positions of erroneous bits. Then we separately train the stacked LSTM models to predict the position of both the first and second erroneous bits and whether to continue flipping. As a result, the DL-aided SCLF decoding algorithms based on the proposed stacked LSTM \mbox{flip-1} model, stacked LSTM \mbox{flip-2} model, and the stacked LSTM \mbox{continue-flipping} check (CFC) model are able to provide a better performance at a lower number of average decoding attempts when compared to other state-of-the-art decoding algorithms.

eess.SP↗

arXiv2026

Informed Dynamic Scheduling for QLDPC Codes

Recent research has shown that syndrome-based belief propagation using layered scheduling (sLBP) can not only accelerate the convergence rate but also improve the error rate performance by breaking the quantum trapping sets for quantum low-density parity-check (QLDPC) codes, showcasing a result distinct from classical error correction codes. In this paper, we consider edge-wise informed dynamic scheduling (IDS) for QLDPC codes based on syndrome-based residual belief propagation (sRBP). However, the construction of QLDPC codes and the identical prior intrinsic information assignment will result in an equal residual in many edges, causing a performance limitation for sRBP. Two heuristic strategies, including edge pool design and error pre-correction, are introduced to tackle this obstacle and quantum trapping sets. Then, a novel sRBP equipped with a predict-and-reduce-error mechanism (PRE-sRBP) is proposed, which can provide over one order of performance gain on the considered bicycle codes and symmetric hypergraph (HP) code under similar iterations compared to sLBP.

quant-ph↗