Researchers have introduced a new metric called Weighted Regret for online multiple testing, aiming to unify the evaluation of false discovery rates and statistical power. They proved a Duality of Regret Conservation, showing that deterministic procedures ensuring FDR control incur significant regret due to threshold depletion. To address this, they proposed Decoupled-OMT (DOMT), a meta-wrapper that uses a decoupled perturbation to reduce regret, especially in bursty environments, while guaranteeing no additional false negatives. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a new metric and algorithm for statistical inference, potentially improving automated pipelines.
RANK_REASON The cluster contains an academic paper detailing a new metric and algorithm for statistical inference. [lever_c_demoted from research: ic=1 ai=0.7]