Informational and computation efficiency of set predictors. / Vovk, Vladimir.

Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering. ed. / Steven de Rooij; Wojciech Kotlowski; Jorma Rissanen; Petri Millimaki; Teemu Roos; Kenji Yamanishi. Amsterdam : CWI, 2012. p. 48-51.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Published

Standard

Informational and computation efficiency of set predictors. / Vovk, Vladimir.

Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering. ed. / Steven de Rooij; Wojciech Kotlowski; Jorma Rissanen; Petri Millimaki; Teemu Roos; Kenji Yamanishi. Amsterdam : CWI, 2012. p. 48-51.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Vovk, V 2012, Informational and computation efficiency of set predictors. in S de Rooij, W Kotlowski, J Rissanen, P Millimaki, TR & K Yamanishi (eds), Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering. CWI, Amsterdam, pp. 48-51.

APA

Vovk, V. (2012). Informational and computation efficiency of set predictors. In S. de Rooij, W. Kotlowski, J. Rissanen, P. Millimaki, T. R., & K. Yamanishi (Eds.), Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering (pp. 48-51). CWI.

Vancouver

Vovk V. Informational and computation efficiency of set predictors. In de Rooij S, Kotlowski W, Rissanen J, Millimaki P, TR, Yamanishi K, editors, Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering. Amsterdam: CWI. 2012. p. 48-51

Author

Vovk, Vladimir. / Informational and computation efficiency of set predictors. Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering. editor / Steven de Rooij ; Wojciech Kotlowski ; Jorma Rissanen ; Petri Millimaki ; Teemu Roos ; Kenji Yamanishi. Amsterdam : CWI, 2012. pp. 48-51

BibTeX

@inproceedings{696f99d95d5348cca3153cbf8dbd53cf,
title = "Informational and computation efficiency of set predictors",
abstract = "There are two methods of set prediction that are provably valid under the assumption of randomness: transductive conformal prediction and inductive conformal prediction. The former method is informationally efficient but often lacks computational efficiency. The latter method is, vice versa, computationally efficient but less efficient informationally. This talk discusses a new method, which we call cross-conformal prediction, that combines informational efficiency of transductive conformal prediction with computational efficiency of inductive conformal prediction. The downside of the new method is that its validity is an empirical rather than mathematical fact.",
author = "Vladimir Vovk",
note = "http://event.cwi.nl/witmse2012/proc.pdf",
year = "2012",
month = sep,
language = "English",
pages = "48--51",
editor = "{de Rooij}, Steven and Wojciech Kotlowski and Jorma Rissanen and Petri Millimaki and {Teemu Roos} and Kenji Yamanishi",
booktitle = "Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering",
publisher = "CWI",

}

RIS

TY - GEN

T1 - Informational and computation efficiency of set predictors

AU - Vovk, Vladimir

N1 - http://event.cwi.nl/witmse2012/proc.pdf

PY - 2012/9

Y1 - 2012/9

N2 - There are two methods of set prediction that are provably valid under the assumption of randomness: transductive conformal prediction and inductive conformal prediction. The former method is informationally efficient but often lacks computational efficiency. The latter method is, vice versa, computationally efficient but less efficient informationally. This talk discusses a new method, which we call cross-conformal prediction, that combines informational efficiency of transductive conformal prediction with computational efficiency of inductive conformal prediction. The downside of the new method is that its validity is an empirical rather than mathematical fact.

AB - There are two methods of set prediction that are provably valid under the assumption of randomness: transductive conformal prediction and inductive conformal prediction. The former method is informationally efficient but often lacks computational efficiency. The latter method is, vice versa, computationally efficient but less efficient informationally. This talk discusses a new method, which we call cross-conformal prediction, that combines informational efficiency of transductive conformal prediction with computational efficiency of inductive conformal prediction. The downside of the new method is that its validity is an empirical rather than mathematical fact.

M3 - Conference contribution

SP - 48

EP - 51

BT - Proceedings of the Fifth Workshop on Information-Theoretic Methods in Science and Engineering

A2 - de Rooij, Steven

A2 - Kotlowski, Wojciech

A2 - Rissanen, Jorma

A2 - Millimaki, Petri

A2 - , Teemu Roos

A2 - Yamanishi, Kenji

PB - CWI

CY - Amsterdam

ER -