Universal prediction

Neri Merhav; Meir Feder

doi:10.1109/18.720534

Universal prediction

Neri Merhav, Meir Feder

School of Electrical Engineering

Research output: Contribution to journal › Article › peer-review

Abstract

This paper consists of an overview on universal prediction from an information-theoretic perspective. Special attention is given to the notion of probability assignment under the self-information loss function, which is directly related to the theory of universal data compression. Both the probabilistic setting and the deterministic setting of the universal prediction problem are described with emphasis on the analogy and the differences between results in the two settings.

Original language	English
Pages (from-to)	2124-2147
Number of pages	24
Journal	IEEE Transactions on Information Theory
Volume	44
Issue number	6
DOIs	https://doi.org/10.1109/18.720534
State	Published - 1998

Keywords

Bayes envelope
Entropy
Finite-state machine
Linear prediction
Loss function
Probability assignment
Redundancy-capacity
Stochastic complexity
Universal coding
Universal prediction

Access to Document

10.1109/18.720534

Cite this

@article{09166e96902f4bde8d619fc0170b6608,

title = "Universal prediction",

abstract = "This paper consists of an overview on universal prediction from an information-theoretic perspective. Special attention is given to the notion of probability assignment under the self-information loss function, which is directly related to the theory of universal data compression. Both the probabilistic setting and the deterministic setting of the universal prediction problem are described with emphasis on the analogy and the differences between results in the two settings.",

keywords = "Bayes envelope, Entropy, Finite-state machine, Linear prediction, Loss function, Probability assignment, Redundancy-capacity, Stochastic complexity, Universal coding, Universal prediction",

author = "Neri Merhav and Meir Feder",

note = "Funding Information: Manuscript received December 1, 1997; revised April 8, 1998. This work was supported by the Israel Science Foundation administered by the Israeli Academy of Sciences and Humanities. N. Merhav is with the Department of Electrical Engineering, Tech-nion–Israel Institute of Technology, Haifa 32000, Israel (e-mail: mer-hav@ee.technion.ac.il). M. Feder is with the Department of Electrical Engineering–Systems, Tel Aviv University, Tel Aviv 69978, Israel (e-mail: meir@eng.tau.ac.il). Publisher Item Identifier S 0018-9448(98)05086-X.",

year = "1998",

doi = "10.1109/18.720534",

language = "אנגלית",

volume = "44",

pages = "2124--2147",

journal = "IEEE Transactions on Information Theory",

issn = "0018-9448",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "6",

}

TY - JOUR

T1 - Universal prediction

AU - Merhav, Neri

AU - Feder, Meir

N1 - Funding Information: Manuscript received December 1, 1997; revised April 8, 1998. This work was supported by the Israel Science Foundation administered by the Israeli Academy of Sciences and Humanities. N. Merhav is with the Department of Electrical Engineering, Tech-nion–Israel Institute of Technology, Haifa 32000, Israel (e-mail: mer-hav@ee.technion.ac.il). M. Feder is with the Department of Electrical Engineering–Systems, Tel Aviv University, Tel Aviv 69978, Israel (e-mail: meir@eng.tau.ac.il). Publisher Item Identifier S 0018-9448(98)05086-X.

PY - 1998

Y1 - 1998

N2 - This paper consists of an overview on universal prediction from an information-theoretic perspective. Special attention is given to the notion of probability assignment under the self-information loss function, which is directly related to the theory of universal data compression. Both the probabilistic setting and the deterministic setting of the universal prediction problem are described with emphasis on the analogy and the differences between results in the two settings.

AB - This paper consists of an overview on universal prediction from an information-theoretic perspective. Special attention is given to the notion of probability assignment under the self-information loss function, which is directly related to the theory of universal data compression. Both the probabilistic setting and the deterministic setting of the universal prediction problem are described with emphasis on the analogy and the differences between results in the two settings.

KW - Bayes envelope

KW - Entropy

KW - Finite-state machine

KW - Linear prediction

KW - Loss function

KW - Probability assignment

KW - Redundancy-capacity

KW - Stochastic complexity

KW - Universal coding

KW - Universal prediction

UR - http://www.scopus.com/inward/record.url?scp=0032184197&partnerID=8YFLogxK

U2 - 10.1109/18.720534

DO - 10.1109/18.720534

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:0032184197

VL - 44

SP - 2124

EP - 2147

JO - IEEE Transactions on Information Theory

JF - IEEE Transactions on Information Theory

SN - 0018-9448

IS - 6

ER -

Universal prediction

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this