Haarnoja et al.: Soft Actor-Critic stochastic off-policy control

What changed

Combined an entropy-regularized stochastic actor with off-policy value/policy updates for sample-efficient continuous control.

Scientific scope

Combined an entropy-regularized stochastic actor with off-policy value/policy updates for sample-efficient continuous control. Maximum-entropy RL and soft Q-learning precede SAC. The original version is not the later automatic-temperature/two-Q implementation in every detail. Empirical seed robustness is not a global convergence or physical-safety guarantee.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Maximum-entropy RL and soft Q-learning precede SAC. The original version is not the later automatic-temperature/two-Q implementation in every detail. Empirical seed robustness is not a global convergence or physical-safety guarantee.

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.

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Haarnoja et al.: Soft Actor-Critic stochastic off-policy control — author/institution implementation

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