BARE is a bandit strategy for identifying the most influential node in an unknown graph, with regret scaling with the detectable dimension rather than the full number of nodes.
Minimax policies for adversarial and stochastic bandits
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
-
Revealing graph bandits for maximizing local influence
BARE is a bandit strategy for identifying the most influential node in an unknown graph, with regret scaling with the detectable dimension rather than the full number of nodes.