Oct 2026
Semibandit Learning for Monotone Stochastic Optimization
Firms can improve decisions as data arrive without pausing operations for a separate data-gathering phase, and even with partial, censored, or yes-or-no feedback, the cumulative performance gap grows essentially at the square root rate.
Rohan Ghuge, Zheng-Jia Zhuo, Arpit Agarwal et al.
· Operational Research · 2 citations
· ⚡1