Well-calibrated predictions from on-line compression models

Authors:
Vladimir Vovk
Affiliations:
Computer Learning Research Centre, Department of Computer Science, Royal Holloway, University of London, Egham, Surrey, UK
Venue:
Theoretical Computer Science - Algorithmic learning theory
Year:
2006

Citing 7
Cited 0

Introduction to Bayesian Networks

Introduction to Bayesian Networks
On-Line Confidence Machines Are Well-Calibrated

FOCS '02 Proceedings of the 43rd Symposium on Foundations of Computer Science
Machine-Learning Applications of Algorithmic Randomness

ICML '99 Proceedings of the Sixteenth International Conference on Machine Learning
Transduction with Confidence and Credibility

IJCAI '99 Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence
Asymptotic Optimality of Transductive Confidence Machine

ALT '02 Proceedings of the 13th International Conference on Algorithmic Learning Theory
Kolmogorov's Contributions to the Foundations of Probability

Problems of Information Transmission
A Universal Well-Calibrated Algorithm for On-line Classification

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

It has been shown recently that transductive confidence machine (TCM) is automatically well-calibrated when used in the on-line mode and provided that the data sequence is generated by an exchangeable distribution. In this paper we strengthen this result by relaxing the assumption of exchangeability of the data-generating distribution to the much weaker assumption that the data agrees with a given "on-line compression model".