Multinomial Thompson sampling for rating scales and prior considerations for calibrating uncertainty

DOI10.1007/S10260-023-00732-YMaRDI QIDQ6580646FDOQ6580646

Authors: Nina Deliu 

Publication date: 29 July 2024

Published in: Statistical Methods and Applications (Search for Journal in Brave)

zbMATH Keywords

Dirichlet distributionmultinomial modelmulti-armed banditsincomplete learningThompson samplingrating scalesadaptive experiments

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• Regression for categorical data.
• Title not available (Why is that?)
• Handbook of Multilevel Analysis
• Bandit algorithms
• Title not available (Why is that?)
• Asymptotically efficient adaptive allocation rules
• Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges
• Active Learning in Multi-armed Bandits
• The class of \textsc{cub} models: statistical foundations, inferential issues and empirical evidence
• Bandit problems with infinitely many arms
• Adaptive treatment assignment in experiments for policy choice
• On incomplete learning and certainty-equivalence control
• On the Prior Sensitivity of Thompson Sampling
• A Tutorial on Thompson Sampling
• Near-optimal regret bounds for Thompson sampling
• Hierarchical marginal models with latent uncertainty
• A response-adaptive randomization procedure for multi-armed clinical trials with normally distributed outcomes
• Response-adaptive randomization in clinical trials: from myths to practical considerations
• On the likelihood that one unkrown probability exeeds another in view of the evidence of two samples.
• Handbook of statistical methods for randomized controlled trials
• Hidden Markov models for longitudinal rating data with dynamic response styles
• Randomization: the forgotten component of the randomized clinical trial

This page was built for publication: Multinomial Thompson sampling for rating scales and prior considerations for calibrating uncertainty

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6580646)