Quantum

How Many Shots Does a Quantum Reservoir Need? Measurement Cost and Its Scaling in Buffered Quantum Reservoir Computing

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 09/11/2026, 05:13:24 UTC.

Briefings

A new arXiv preprint argues that quantum reservoir computing requires two distinct measurement budgets: a precision budget to estimate features to a fixed accuracy, and a convergence budget to reach the noise-free ceiling. The paper demonstrates these budgets are not equivalent, meaning configurations with worse noise floors are cheaper to optimize, necessitating fixed budget comparisons rather than budget-dependent ones.

Key points

• Quantum reservoir computing requires two distinct measurement budgets: a precision budget to estimate features to a fixed accuracy, and a convergence budget to reach the noise-free ceiling.
• Proposition 1 shows that convergence budget depends on the noise-free error floor by construction, making configurations with worse noise floors appear cheaper to optimize.

Context

Quantum reservoir computing inherits from classical reservoir computing the property that only a linear readout is trained, but unlike a classical reservoir its state is never directly observable: every feature is an expectation value estimated from a finite number of measurement shots. The shot budget is a first-class hyperparameter with no classical analogue. This paper distinguishes a precision budget from a convergence budget to resolve this uncertainty.

Limitations

The evidence is an arXiv preprint (arXiv:2611.08632v1) from November 2026, meaning it represents unverified research. The paper explicitly states that competing definitions of the required budget do not agree, and the systematic disagreement is not fully resolved. Specific numerical values for required shot counts are not provided in the abstract.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 09/11/2026, 07:18:01 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeic7ob4e7rtkqf7zr6pouocw2a5uulgalbk3atikaznckalo5evoda