Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Offline Safe Reinforcement Learning research benchmarks · SOTA2 Research

Offline Safe Reinforcement Learning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.74Normalized Reward

5

Mar 25, 2026

Bullet-SafetyGym

68Normalized Reward

5

Mar 25, 2026

0.54Normalized Reward

5

Mar 25, 2026

270.45Reward

4

Jul 7, 2026

243.63Reward

4

Jul 7, 2026

253.58Reward

4

Jul 7, 2026

DSRL HardSparse

257.98Reward

4

Jul 7, 2026

DSRL MediumDense

237.75Reward

4

Jul 7, 2026

DSRL MediumMean

236.57Reward

4

Jul 7, 2026

DSRL MediumSparse

242.75Reward

4

Jul 7, 2026

316.47Reward

4

Jul 7, 2026

DSRL EasySparse

326.06Reward

4

Jul 7, 2026

Popular datasets

DSRL MetaDrive SafetyGym Bullet-SafetyGym

Follow for updates

Join SOTA2 Contact