Dominant Arm Identification with Mixing and Recycling Observed Samples

Published in Preprint, 2026

Research Presentation

Download Full PDF
Title slide for Dominant Arm Identification with Mixing and Recycling Observed Samples
Slide explaining how crossing reward distributions can change the selected arm
Slide explaining how DRDSE learns every reward CDF from partial feedback
Slide comparing the upper and lower sample-complexity bounds
Slide showing DRDSE sample-complexity results in Gaussian and biomedical settings