AWhat changed
Added a quadratic penalty around the broadcast model to local client objectives, allowing inexact local solutions while analyzing progress under a device-dissimilarity condition.
Scientific scope
Added a quadratic penalty around the broadcast model to local client objectives, allowing inexact local solutions while analyzing progress under a device-dissimilarity condition. Original 2018 front starts Anit Kumar Sahu then Tian Li, equal-contribution marked; later title/order not retroactively substituted. Inherits proximal regularization and FedAvg. Theory needs bounded local dissimilarity away from approximate stationarity and specified smoothness/solver accuracy; not convergence under arbitrary heterogeneous data, any proximal coefficient or unrestricted inexact solves.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Original 2018 front starts Anit Kumar Sahu then Tian Li, equal-contribution marked; later title/order not retroactively substituted. Inherits proximal regularization and FedAvg. Theory needs bounded local dissimilarity away from approximate stationarity and specified smoothness/solver accuracy; not convergence under arbitrary heterogeneous data, any proximal coefficient or unrestricted inexact solves.
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.
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