Dynamic Algorithm Configuration on OneMax n=1000 v1
6.017ERTpi_opt
Evaluation results
| Method | Links | ||
|---|---|---|---|
| pi_opttype=optimal policy2025.12 | 6.017 | — | |
| irace2025.12 | 6.422 | 20,046 | |
| DDQN with adaptive reward shiftingReward function=Adaptive Shifted, gamma=0.992025.12 | 6.491 | 0.032 | |
| pi_conttype=theory-derived continuous policy2025.12 | 6.587 | — | |
| pi_disctype=theory-derived discrete policy2025.12 | 6.701 | — | |
| DDQNReward function=Naïve, gamma=12025.12 | 6.72 | 0.476 | |
| DDQNReward function=Scaled, gamma=12025.12 | 6.734 | 0.024 |