List your company
ResearchFormerly WizWand

Lifelong Multi-agent Pathfinding on Lifelong MAPF Map (train)

2,248.71Cumulative Original RewardIPPO (ARMS)
703.10361,104.36681,505.631,906.8932May 22, 2026
Updated 5mo ago

Evaluation results

MethodLinks
IPPO (ARMS)Agent Count=16, Reward Sparsity=10, Base RL Algorithm=IPPO2026.052,248.71
IPPO (ARMS)Agent Count=16, Reward Sparsity=20, Base RL Algorithm=IPPO2026.052,155.95
IPPO (ARMS)Agent Count=8, Reward Sparsity=20, Base RL Algorithm=IPPO2026.052,116.54
IPPO (ARMS)Agent Count=32, Reward Sparsity=20, Base RL Algorithm=IPPO2026.052,071.69
IPPO (ARMS)Agent Count=16, Reward Sparsity=30, Base RL Algorithm=IPPO2026.051,799.66
IPPO (No shaping)Agent Count=16, Reward Sparsity=10, Base RL Algorithm=IPPO2026.051,619.27
IPPO (PBRS)Agent Count=16, Reward Sparsity=10, Base RL Algorithm=IPPO2026.051,600.68
IPPO (PBRS)Agent Count=8, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,413.82
IPPO (PBRS)Agent Count=16, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,306.5
IPPO (No shaping)Agent Count=16, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,212.87
IPPO (No shaping)Agent Count=8, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,114.58
IPPO (PBRS)Agent Count=32, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,076.61
IPPO (No shaping)Agent Count=32, Reward Sparsity=20, Base RL Algorithm=IPPO2026.051,022.83
IPPO (PBRS)Agent Count=16, Reward Sparsity=30, Base RL Algorithm=IPPO2026.05904.65
IPPO (No shaping)Agent Count=16, Reward Sparsity=30, Base RL Algorithm=IPPO2026.05762.55