The paper formalizes hidden Byzantine action overwrites in cooperative multi-agent RL, proves an exact rectangular robust-MDP reduction, shows linear security regret is unavoidable under uninformative attacks, and gives a stage-tied E2D learner with ~O(H^2 S sqrt(AK)) + E[D_K] regret.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Online Security Learning in Cooperative Multi-Agent Systems under Hidden Byzantine Attacks
The paper formalizes hidden Byzantine action overwrites in cooperative multi-agent RL, proves an exact rectangular robust-MDP reduction, shows linear security regret is unavoidable under uninformative attacks, and gives a stage-tied E2D learner with ~O(H^2 S sqrt(AK)) + E[D_K] regret.