ResearchTasksPolicy Gradient OptimizationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedFirst-order stationary point convergence (||∇J(θ)||^2 ≤ ε)REINFORCE1Memory Usage (#Trajectories)8Jun 4, 2026Unregularized Objective Tabular RL——Primary metric0Feb 26, 2026
First-order stationary point convergence (||∇J(θ)||^2 ≤ ε)REINFORCE1Memory Usage (#Trajectories)8Jun 4, 2026