Inverted Pendulum
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Safety Verification | Inverted Pendulum Adversarial Training | Success Rate99.3 | 12 | |
| Safety Verification | Inverted Pendulum (subregions certifiable) | Success Rate (%)0.911 | 12 | |
| Control Task | Inverted Pendulum (test) | Average Reward996.35 | 6 | |
| Inverted Pendulum Control | Inverted Pendulum (test) | ROA6,131 | 6 | |
| Trajectory Optimization | Inverted Pendulum | Runtime (seconds)1.86 | 5 | |
| Reinforcement Learning | Inverted Pendulum | Maximum Evaluation Return1,000 | 5 | |
| Inverted Pendulum Balancing | Inverted Pendulum 10 initial states | Time (ms)0.038 | 4 | |
| Lyapunov stability verification | Inverted Pendulum large torque | Time (s)1.8 | 4 | |
| Inverted Pendulum Balancing | Inverted Pendulum | 95% Convergence Steps32.4 | 3 | |
| Lyapunov stability verification | Inverted Pendulum output | Time (s)5.3 | 3 | |
| Lyapunov stability verification | Inverted Pendulum small torque | Time (s)2.2 | 3 | |
| Inverted Pendulum | Inverted Pendulum 10 challenging instances | Average Steps500 | 3 | |
| Continuous Control | Inverted Pendulum | Normalized AUC0.97 | 3 | |
| System Verification and Training | Inverted Pendulum output | Training Time (min)7.3 | 2 | |
| System Verification and Training | Inverted Pendulum | Training Time (min)1.7 | 2 | |
| Sensory-motor control | Inverted Pendulum MuJoCo | Mean Best Reward829.66 | 2 | |
| PDF approximation and error bound estimation of SDE states | 2D Inverted Pendulum | Time (p)45,409 | 1 |
Showing 17 of 17 rows