ResearchBenchmarksSafe Reinforcement Learning on Safety Gymnasium HalfCheetahFollow3.206Time To Safety (TTS)APPO1.9049210.6872119.469528.25179Jun 12, 2026Evaluation ResultsMethodMethodLinksTime To Safety (TTS)Risk Profile (RP)Number of Violations (#V)APPO2026.063.2060.998683.8CSPO2026.068.2291.00340.25CPPOPID2026.069.3231.0018118.4PPO-Lag2026.0626.291.0075110.2CUP2026.0630.5331.020290.2FOCOPS2026.0635.7331.0249116.88