Scalable evolution-strategy training of deep control policies

What changed

Demonstrated parallel natural-evolution-strategy style optimization of deep policy parameters using shared random perturbations and scalar fitness communication.

Scientific scope

Demonstrated parallel natural-evolution-strategy style optimization of deep policy parameters using shared random perturbations and scalar fitness communication. Evolution strategies, NES and score-function gradient estimators predate this work; credited contribution is the scalable deep-policy implementation and empirical study. Wall-clock results use hundreds/thousands of CPUs and often more environment samples than comparators.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Evolution strategies, NES and score-function gradient estimators predate this work; credited contribution is the scalable deep-policy implementation and empirical study. Wall-clock results use hundreds/thousands of CPUs and often more environment samples than comparators.

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.

• Evolution Strategies as a Scalable Alternative to Reinforcement Learning

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