WRSWorld Research StateContribute
WORLD RESEARCH STATE

FTRL-Proximal: cumulative nonsmooth penalties and sparse online prediction

Contribution grade

Regularized online learning and industrial prediction · Human-designed research method or resource

Reference score175,000Provisional editorial estimate of this exact contribution; its grade is assessed separately.

What changed

Related mirror-descent and FTRL updates, retaining cumulative nonsmooth penalties with proximal centering; later industrial work implemented sparse per-coordinate online logistic prediction and calibration/feature-management lessons.

Scientific scope

Related mirror-descent and FTRL updates, retaining cumulative nonsmooth penalties with proximal centering; later industrial work implemented sparse per-coordinate online logistic prediction and calibration/feature-management lessons. 2011single-author theory and2013sixteen-author industrial case study kept as source-specific stages, not sixteen inventors of2011algorithm. Inherits earlier FTRL/dual-averaging/proximal work including2010adaptive-bound precursor. Equivalence claims require specified regularization/update assumptions; sparse click prediction is not causal ad effectiveness or universal superiority.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. 2011single-author theory and2013sixteen-author industrial case study kept as source-specific stages, not sixteen inventors of2011algorithm. Inherits earlier FTRL/dual-averaging/proximal work including2010adaptive-bound precursor. Equivalence claims require specified regularization/update assumptions; sparse click prediction is not causal ad effectiveness or universal superiority.

Contributors

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