On-line maximum likelihood prediction with respect to general loss functions

DOI10.1006/jcss.1997.1503zbMath0880.68053MaRDI QIDQ1370862

Kenji Yamanishi

Publication date: 28 October 1997

Published in: Journal of Computer and System Sciences (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1006/jcss.1997.1503

zbMATH Keywords

on-line prediction algorithms; parametric probabilistic models

Mathematics Subject Classification ID

68W10: Parallel algorithms in computer science

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• A universal prior for integers and estimation by minimum description length
• A learning criterion for stochastic rules
• The weighted majority algorithm
• Efficient distribution-free learning of probabilistic concepts
• A loss bound model for on-line stochastic prediction algorithms
• Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach
• Information-theoretic asymptotics of Bayes methods
• A universal data compression system
• Universal coding, information, prediction, and estimation
• Best ?- and N ?-approximants in Orlicz spaces of vector valued functions
• Minimum complexity density estimation
• The strong law of large numbers for sequential decisions under uncertainty
• On the Assumptions Used to Prove Asymptotic Normality of Maximum Likelihood Estimates