VGM²P achieves SOTA-comparable performance in offline MARL via value-guided conditional behavior cloning with MeanFlow, enabling efficient single-step action generation insensitive to regularization coefficients.
Facmac: Factored multi-agent centralised policy gradients.Advances in Neural Information Pro- cessing Systems, 34:12208–12221, 2021
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Value-Guidance MeanFlow for Offline Multi-Agent Reinforcement Learning
VGM²P achieves SOTA-comparable performance in offline MARL via value-guided conditional behavior cloning with MeanFlow, enabling efficient single-step action generation insensitive to regularization coefficients.