NAS-Bench-301 learned surrogate architecture benchmark

What changed

Fits predictive surrogate benchmarks using evaluated DARTS-space architectures to simulate searches over a much larger space than exhaustive tabulation can cover.

Scientific scope

Fits predictive surrogate benchmarks using evaluated DARTS-space architectures to simulate searches over a much larger space than exhaustive tabulation can cover. The approximately 10^18 space is modeled from tens of thousands of evaluated candidates, not exhaustively trained. Surrogate error, sampling bias and training/noise assumptions remain; its predictions are not exact measurements or proof of a search algorithm's real-world ranking. First two authors contribute equally.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The approximately 10^18 space is modeled from tens of thousands of evaluated candidates, not exhaustively trained. Surrogate error, sampling bias and training/noise assumptions remain; its predictions are not exact measurements or proof of a search algorithm's real-world ranking. First two authors contribute equally.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• NAS-Bench-301 and the Case for Surrogate Benchmarks for Neural Architecture Search

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction