Heterogeneous Pendulum
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Heterogeneous Pendulum Rich-Data 600,000 transition steps
-0.6Cumulative Reward
7
Heterogeneous Pendulum 300,000 transition steps (Mid-Data)
-1.25Cumulative Reward
7
Heterogeneous Pendulum Low-Data 100,000 transition steps
-1.39Cumulative Reward
7