ResearchDatasetsQ*bertQ*bertFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsBackdoor Attack on Reinforcement LearningQ*bert Discrete (evaluation)18,052BR5Reinforcement LearningQ*bert Atari 2600 (test)18,900Average Total Reward5Deep Reinforcement LearningQ*bert Atari 26008,160IQM Return4Real-time game playingQ*bert185Score3