Abstract
Purchase PDF (182 K)
E-mail Article
Add to my Quick Links
Purchase PDF (300 K)
doi:10.1016/j.tcs.2006.07.037 
Copyright © 2006 Elsevier B.V. All rights reserved.
Criterion of calibration for transductive confidence machine with limited feedback
Available online 31 July 2006.
References and further reading may be available for this article. To view references and further reading you must purchase this article.
Abstract
This paper is concerned with the problem of on-line prediction in the situation where some data are unlabelled and can never be used for prediction, and even when the data are labelled, the labels may arrive with a delay. We construct a modification of randomised transductive confidence machine for this case and prove a necessary and sufficient condition for its predictions being calibrated, in the sense that in the long run they are wrong with a prespecified probability under the assumption that the data are generated independently by the same distribution. The condition for calibration turns out to be very weak: feedback should be given on more than a logarithmic fraction of steps.
Keywords: On-line prediction; Confident prediction; Transductive confidence machine
