Reinforcement Learning on CartPole (e1, e2, Av. Reward)
6E1Genetic Curriculum Learning
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Genetic Curriculum LearningPhase=22026.06 | 6 | 24 | 93 | |
| Scenario Generation for Risk-Aware Reinforcement LearningPhase=22026.06 | 10 | 15 | 96 | |
| Curriculum Adversarial TrainingPhase=22026.06 | 10 | 19 | 95 | |
| Perturbed ExplorationPhase=22026.06 | 12 | 51 | 62 | |
| CoDEPhase=22026.06 | 15 | 29 | 97 | |
| Epsilon Greedy PPOPhase=22026.06 | 19 | 52 | 83 | |
| Scenario Generation for Risk-Aware Reinforcement LearningPhase=12026.06 | 21 | 48 | 80 | |
| Vanilla PPOPhase=22026.06 | 21 | 43 | 91 |