#confidence intervals
7 papers · 1 filter
Agents That Certify Their Own Exploits: Confidence-Scheduled Restricted Responses for Safe Opponent Exploitation
Boning Li, Longbo Huang
The paper introduces confidence‑scheduled restricted responses (CS‑RNR), a method that lets an agent safely exploit a flawed opponent in two‑player zero‑sum imperfect‑information g…
Edgington's Combination Method for Two-Study Meta-Analysis: An Empirical Evaluation in 1226 Meta-Analyses
Samuel Pawel, Saverio Fontana, Jinyu Chen +6
The paper empirically evaluates Edgington's p‑value combination method as an alternative for two‑study meta‑analyses, showing it adapts confidence‑interval width without needing ex…
Penalized likelihood inference for beta-binomial meta-analysis of proportions of rare events
Kotaro Sasaki, Hisashi Noma
The paper introduces a penalized likelihood estimator using a Jeffreys prior for beta-binomial meta‑analysis of rare event proportions, improving bias and stability over the standa…
Statistical Inference for Scenario-Based Dynamic Optimization under Uncertainty
Aurya Javeed, Johannes Milz
The paper develops statistical methods for evaluating and constructing confidence intervals for the optimal value of finite-horizon open-loop dynamic optimization problems when par…
How segmented is my network?
Rohit Dube
The paper introduces a statistically grounded metric called segmentedness to quantify how much a network is segmented, provides a normalized estimator with confidence intervals, an…
Efficient Conformal Prediction for Regression Models under Label Noise
Yahav Cohen, Jacob Goldberger, Tom Tirer
The paper introduces a method to apply conformal prediction to regression models when the calibration data have noisy labels, by estimating a noise‑free threshold and providing a p…