RL policy training interval optimization on Indoor Lighting Trace Stair Access (simulated 1000 nodes)
89Policy Trainings CountBaseline
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Baseline2018.11 | 89 | 0 | — | |
| Dynamic On-Policy2018.11 | 20 | 0 | 77 |
| Method | Links | |||
|---|---|---|---|---|
| Baseline2018.11 | 89 | 0 | — | |
| Dynamic On-Policy2018.11 | 20 | 0 | 77 |