Reinforcement Learning on CartPole Pure
200Average Reward (2/0.5)True env+MPC
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| True env+MPCData source (optimality)=Oracle (True Environment)2021.02 | 200 | 198.4 | 200 | |
| True env+MPCData source (optimality)=Oracle (True Environment)2021.02 | 200 | 198.4 | 200 | |
| True env+MPCData source (optimality)=Oracle (True Environment)2021.02 | 200 | 198.4 | 200 | |
| PerSimData source (optimality)=Pure-8-202021.02 | 199.8 | 199.1 | 197.8 | |
| PerSimData source (optimality)=Pure2021.02 | 199.7 | 193.8 | 192 | |
| PerSimData source (optimality)=Random2021.02 | 197.7 | 193 | 185.7 | |
| Vanilla CaDMData source (optimality)=Pure-8-202021.02 | 171.1 | 193.4 | 64.2 | |
| Vanilla CaDMData source (optimality)=Pure2021.02 | 168 | 190.8 | 58.1 | |
| MORELData source (optimality)=Pure-8-20, Algorithm variant=Planning2021.02 | 166.7 | 197.7 | 189.4 | |
| BCQData source (optimality)=Pure, Algorithm variant=Planning2021.02 | 166.2 | 181.2 | 182.8 | |
| CQLData source (optimality)=Pure-8-20, Algorithm variant=Planning2021.02 | 163.7 | 198.1 | 190.4 | |
| CQLData source (optimality)=Pure, Algorithm variant=Planning2021.02 | 154.4 | 190.9 | 170.2 | |
| Vanilla CaDMData source (optimality)=Random2021.02 | 150.5 | 175.6 | 65.1 | |
| CQLData source (optimality)=Pure, Algorithm variant=Acting2021.02 | 122.8 | 179.4 | 193.6 | |
| BCQData source (optimality)=Pure-8-20, Algorithm variant=Planning2021.02 | 98.9 | 162.1 | 86.1 | |
| PE-TS CaDMData source (optimality)=Pure-8-202021.02 | 98.3 | 198.6 | 141.1 | |
| PE-TS CaDMData source (optimality)=Pure2021.02 | 92.3 | 193.6 | 127.5 | |
| PE-TS CaDMData source (optimality)=Random2021.02 | 88.6 | 196.1 | 171 | |
| CQLData source (optimality)=Random, Algorithm variant=Acting2021.02 | 67.4 | 30.6 | 7 | |
| BCQData source (optimality)=Pure-8-20, Algorithm variant=Acting2021.02 | 67.3 | 65.6 | 140 | |
| BCQData source (optimality)=Pure, Algorithm variant=Acting2021.02 | 65.4 | 79.2 | 132.1 | |
| MORELData source (optimality)=Random, Algorithm variant=Planning2021.02 | 50.2 | 68.9 | 40.4 | |
| BCQData source (optimality)=Random, Algorithm variant=Planning2021.02 | 44.8 | 57.91 | 36.4 | |
| BCQData source (optimality)=Random, Algorithm variant=Acting2021.02 | 43.9 | 21.1 | 39.5 | |
| CQLData source (optimality)=Pure-8-20, Algorithm variant=Acting2021.02 | 42.4 | 199 | 199.8 | |
| MORELData source (optimality)=Pure-8-20, Algorithm variant=Acting2021.02 | 41.4 | 198.8 | 199.9 | |
| CQLData source (optimality)=Random, Algorithm variant=Planning2021.02 | 39.9 | 77.9 | 148.7 | |
| MORELData source (optimality)=Pure, Algorithm variant=Planning2021.02 | 35.8 | 96.6 | 66.4 | |
| MORELData source (optimality)=Random, Algorithm variant=Acting2021.02 | 35.8 | 27.5 | 10.6 | |
| MORELData source (optimality)=Pure, Algorithm variant=Acting2021.02 | 33.7 | 27.5 | 10.1 |