The paper introduces the PUCS framework with greedy probing and an online algorithm OLPA, claiming O(√T + ln²T) regret, but the regret proof relies on an unjustified optimality assumption.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Online Learning with Probing for Sequential User-Centric Selection
The paper introduces the PUCS framework with greedy probing and an online algorithm OLPA, claiming O(√T + ln²T) regret, but the regret proof relies on an unjustified optimality assumption.