Dueling networks: separate state-value and action-advantage streams

What changed

Factored a deep action-value architecture into shared features and separate state-value/action-advantage streams, enabling learning across actions with similar values.

Scientific scope

Factored a deep action-value architecture into shared features and separate state-value/action-advantage streams, enabling learning across actions with similar values. Inherits advantage-learning ideas and DQN; the contribution is the neural factoring architecture, not invention of advantage functions or guaranteed improvement for all action spaces.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits advantage-learning ideas and DQN; the contribution is the neural factoring architecture, not invention of advantage functions or guaranteed improvement for all action spaces.

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.

• Dueling Network Architectures for Deep 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