Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Boxpushing

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-agent Reinforcement LearningBoxpushing
Reward145.35
15
Multi-agent coordinationBoxpushing (BP)
Base Score290.42
4
RNN-based cooperative multi-agent verificationBoxPushing (BP) 20x20 environment
Avg Violation Rate1.51
2
RNN-based cooperative multi-agent verificationBoxPushing (BP) 10x10 environment
Average Violation Rate1.15
2
Showing 4 of 4 rows