MAGPIE converts agent-specific preference feedback into per-agent reward models and a summed global reward, with a claimed epsilon-Nash equilibrium guarantee.
Value-decomposition networks for cooperative multi-agent learning based on team reward,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2026 1verdicts
REJECT 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Multi-Agent Reinforcement Learning via Agent-Specific Preference
MAGPIE converts agent-specific preference feedback into per-agent reward models and a summed global reward, with a claimed epsilon-Nash equilibrium guarantee.