A neural decoder that identifies quantum errors more accurately than hand-designed methods
A transformer trained on Google's Sycamore surface-code data cut decoding errors by 6% against tensor-network decoding and 30% against correlated matching, the best reported accuracy at the time.
- Model
- AlphaQubit
- Field
- Physics
- Date
- 2024-11-20
Sources
Original work
Announcement
What was found
A quantum error-correcting code produces a stream of syndrome measurements; a decoder has to infer from them which errors actually occurred. Hand-designed decoders assume a simplified noise model, whereas AlphaQubit is a transformer trained first on simulated data and then fine-tuned on hundreds of millions of real syndrome samples from Sycamore, so it can learn the device's actual correlated noise, leakage and crosstalk. It beat both the accuracy-oriented tensor-network decoder and the fast matching decoder, and held up in simulation to distance-11 codes. Published in Nature.
Novelty check
Machine-learning decoders had been proposed for years but had not beaten the best classical decoders on real hardware data at scale. The result is a decoder that does, on a specific processor.
Caveats and known objections
Accuracy, not speed: AlphaQubit is far slower than matching decoders and does not run inside the real-time budget a fault-tolerant machine needs, which the paper states. It is trained per device on that device's data, so it does not transfer for free. Scaling behaviour beyond the tested code distances is open. This improves a component of error correction; it does not itself demonstrate a fault-tolerant computation.
Nobody outside the lab has checked this yet.
Reading the primary source closely enough to say whether it supports the claim counts as a check, and you are credited on the entry.
Or on GitHub: submit a check challenge the grade send a correction or send a pull request
Flag this for triage
Signals order the review queue and nothing else. They are never published, and they never move a grade: that takes a citation.
Entry history (1 event)
- AddedEntered the registry graded Peer reviewed and AI-led.
Entries are never deleted. A grade that does not hold up is downgraded on the record, with the reason beside it.
Graded peer reviewed for verification and ai-led for autonomy. What these mean.
Cite this entry
whataifound.org. (2024). A neural decoder that identifies quantum errors more accurately than hand-designed methods. whataifound.org: A Registry of AI Scientific and Mathematical Discoveries. https://whataifound.org/finding/2024-11-20-alphaqubitBibTeX
@misc{whataifound-googledeepmind-2024-alphaqubit,
title = {A neural decoder that identifies quantum errors more accurately than hand-designed methods},
author = {{whataifound.org}},
year = {2024},
howpublished = {whataifound.org: A Registry of AI Scientific and Mathematical Discoveries},
note = {Result by Google DeepMind / Google Quantum AI. Verification: Peer reviewed. Autonomy: AI-led.},
url = {https://whataifound.org/finding/2024-11-20-alphaqubit}
}