ResearchTasksAtari game reinforcement learningAtari game reinforcement learningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAtari Seaquest Unexpected Environment (evaluation)PPO + Repaired Shield75.6Reward5Feb 26, 2026Atari Seaquest Expected Environment (train)PPO + Naive Shield377.21Training Reward5Feb 26, 2026