The POW index policy for restless multi-armed bandits with per-arm penalty constraints is asymptotically optimal, computable offline per user, and learnable via deep RL.
Title resolution pending
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
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Restless Bandits with Individual Penalty Constraints: Near-Optimal Indices and Deep Reinforcement Learning
The POW index policy for restless multi-armed bandits with per-arm penalty constraints is asymptotically optimal, computable offline per user, and learnable via deep RL.