REVIEW 8 cited by
Learning to Decode the Surface Code with a Recurrent, Transformer-Based Neural Network
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Quantum error-correction is a prerequisite for reliable quantum computation. Towards this goal, we present a recurrent, transformer-based neural network which learns to decode the surface code, the leading quantum error-correction code. Our decoder outperforms state-of-the-art algorithmic decoders on real-world data from Google's Sycamore quantum processor for distance 3 and 5 surface codes. On distances up to 11, the decoder maintains its advantage on simulated data with realistic noise including cross-talk, leakage, and analog readout signals, and sustains its accuracy far beyond the 25 cycles it was trained on. Our work illustrates the ability of machine learning to go beyond human-designed algorithms by learning from data directly, highlighting machine learning as a strong contender for decoding in quantum computers.
Forward citations
Cited by 8 Pith papers
-
Proof of a finite threshold for the union-find decoder
Union-find decoder for surface code achieves finite threshold under circuit-level stochastic errors with quasi-polylog parallel runtime bound.
-
Physics-Informed Graph-Neural Decoding of the Surface Code: the Logical Signal as an Exact Topological Pairing
The logical-error signal in a surface-code decoder is an exact relative-cohomology pairing of the syndrome with a boundary-fixed harmonic coordinate, evaluated as the current difference between two boundary sinks.
-
Coset Ensemble Decoder for Quantum Error Correction with Algorithm-Hardware Co-Design
Presents a coset ensemble decoder with algorithm-hardware co-design that claims better accuracy-latency trade-off and lower FPGA resource use than MWPM and UF baselines under depolarizing noise.
-
Low Latency GNN Accelerator for Quantum Error Correction
Hardware-guided pruning and quantization make a GNN surface-code decoder meet ~1 µs real-time latency on FPGA while cutting logical error rate 13–40% versus MWPM at d≤7, p=10⁻³.
-
Learning Neural Decoding with Parallelism and Self-Coordination for Quantum Error Correction
A transformer-based decoder trained on local window labels learns to output per-window logical corrections that can be XORed across sliding windows, enabling parallel decoding with accuracy slightly above belief match...
-
Low Latency GNN Accelerator for Quantum Error Correction
An FPGA-accelerated GNN decoder for surface-code quantum error correction delivers sub-1us latency and lower error rates than state-of-the-art approaches for code distances up to 7.
-
Soft information decoding with superconducting qubits
Soft decoding with analog measurement data raises repetition-code thresholds by 25% and reduces error rates up to 30x on superconducting qubits, with one byte per shot sufficient for near-optimal performance.
-
Managing Classical Processing Requirements for Quantum Error Correction
A two-level decoder scheduling framework reduces classical processing requirements for quantum error correction by 10-40% on fault-tolerant benchmarks by managing bursty workloads as shared resources.
Discussion (0). Sign in to comment.