A modular PS-RMAB framework integrates base algorithms with change detection and diminishing exploration to achieve tilde-O(sqrt(L M K T)) excess regret against a segment-oracle benchmark, with simulations showing near-oracle performance.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IT 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Modularized Framework for Piecewise-Stationary Restless Bandits
A modular PS-RMAB framework integrates base algorithms with change detection and diminishing exploration to achieve tilde-O(sqrt(L M K T)) excess regret against a segment-oracle benchmark, with simulations showing near-oracle performance.