List your company

ResearchFormerly WizWand

Point-Maze

Benchmarks

TaskDatasetSOTA resultTrendResults

Inverse Reinforcement Learning

Point Maze

1.03Normalized Performance

6

Policy Generalization

Point Maze (test)

-5.21Average Return

6

Reward Adaptation

Point-Maze Shift (meta-test)

-5.37Average Return

4

Inverse Reinforcement Learning

Point Maze Flipped

96Normalized Performance

3

Policy Generalization

Point-Maze-Shift (meta-test)

-28.61Average Return

3