LTL-guided Reinforcement Learning on Zones Finite-horizon (test)
98Success RateDEEPLTL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| DEEPLTLTask Specification Type=LOCAL-SAFETY[3, 3], |Q|=83, |δ|=3792026.02 | 98 | — | |
| SEMLTLTask Specification Type=FINITE-REACTIVE[8, 2], |Q|=107, |δ|=6192026.02 | 98 | 2.96 | |
| DEEPLTLTask Specification Type=SMALL, |Q|=9, |δ|=232026.02 | 93 | — | |
| SEMLTLTask Specification Type=LOCAL-SAFETY[3, 3], |Q|=83, |δ|=3792026.02 | 93 | 5.09 | |
| SEMLTLTask Specification Type=SMALL, |Q|=9, |δ|=232026.02 | 92 | 3.4 | |
| SEMLTLTask Specification Type=GLOBAL-SAFETY[4, 8], |Q|=1, 162, |δ|=6, 5102026.02 | 79 | 6.73 | |
| LTL2ACTIONTask Specification Type=FINITE-REACTIVE[8, 2], |Q|=107, |δ|=6192026.02 | 74 | — | |
| LTL2ACTIONTask Specification Type=LOCAL-SAFETY[3, 3], |Q|=83, |δ|=3792026.02 | 66 | — | |
| LTL2ACTIONTask Specification Type=SMALL, |Q|=9, |δ|=232026.02 | 52 | — | |
| LTL2ACTIONTask Specification Type=GLOBAL-SAFETY[4, 8], |Q|=1, 162, |δ|=6, 5102026.02 | 39 | — |