Cooperative Multi-Agent Reinforcement Learning on Heterogeneous Cloud Scheduling 40 scenarios (test)
-40.5Best Checkpoint RewardDG-PG
Evaluation results
| Method | Links | |
|---|---|---|
| DG-PGNumber of Agents (N)=202026.02 | -40.5 | |
| Best-FitNumber of Agents (N)=1002026.02 | -40.7 | |
| Best-FitNumber of Agents (N)=2002026.02 | -41 | |
| Best-FitNumber of Agents (N)=502026.02 | -41.1 | |
| DG-PGNumber of Agents (N)=1002026.02 | -41.5 | |
| DG-PGNumber of Agents (N)=502026.02 | -41.7 | |
| Best-FitNumber of Agents (N)=202026.02 | -41.9 | |
| DG-PGNumber of Agents (N)=102026.02 | -43.5 | |
| Best-FitNumber of Agents (N)=102026.02 | -43.7 | |
| DG-PGNumber of Agents (N)=52026.02 | -44.9 | |
| Best-FitNumber of Agents (N)=52026.02 | -46.3 | |
| DG-PGNumber of Agents (N)=2002026.02 | -46.4 | |
| RandomNumber of Agents (N)=2002026.02 | -60 | |
| RandomNumber of Agents (N)=1002026.02 | -60.5 | |
| RandomNumber of Agents (N)=502026.02 | -66.8 | |
| RandomNumber of Agents (N)=102026.02 | -78.7 | |
| RandomNumber of Agents (N)=202026.02 | -80 | |
| RandomNumber of Agents (N)=52026.02 | -84.2 |