We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Philip S Thomas, Erik Learned-Miller and My Phan's contribution to the Discussion of 'Estimating means of bounded random variables by betting' by Waudby-Smith and Ramdas.
- Authors
Thomas, Philip S; Learned-Miller, Erik; Phan, My
- Abstract
The article discusses the importance of developing tighter confidence intervals and sequences for the mean in machine learning systems. The authors argue that providing guarantees in the form of confidence intervals can ensure the safety, fairness, and reliability of these systems. They highlight the significance of confidence intervals that do not rely on normality assumptions and remain relatively tight, especially when dealing with extreme distributions. The authors appreciate the comparison of their own confidence interval to the Hedged-CI method presented in the paper and find the increase in tightness compelling. They also suggest further comparisons with other confidence intervals proposed by Gaffke. The article concludes by providing contact information for correspondence with the authors and stating that there are no conflicts of interest.
- Subjects
RANDOM variables; NONPARAMETRIC statistics
- Publication
Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2024, Vol 86, Issue 1, p51
- ISSN
1369-7412
- Publication type
Article
- DOI
10.1093/jrsssb/qkad124