Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples

Matthias Ernst; Maciej Liskiewicz; Rüdiger Reischuk

doi:10.1007/978-3-662-48971-0_14

Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples

Matthias Ernst, Maciej Liskiewicz, Rüdiger Reischuk

Institut für Theoretische Informatik

Graduate School for Computing in Medicine and Live Science

Abstract

Proper learning from positive samples is a basic ingredient for designing secure steganographic systems for unknown covertext channels. In addition, security requirements imply that the hypothesis should not contain false positives. We present such a learner for k-term DNF formulas for the uniform distribution and a generalization to q-bounded distributions. We briefly also describe how these results can be used to design a secure stegosystem.

Originalsprache	Englisch
Titel	Algorithms and Computation
Redakteure/-innen	Khaled Elbassioni, Kazuhisa Makino
Seitenumfang	12
Band	9472
Herausgeber (Verlag)	Springer Verlag
Erscheinungsdatum	27.11.2015
Seiten	151-162
ISBN (Print)	978-3-662-48970-3
ISBN (elektronisch)	978-3-662-48971-0
DOIs	https://doi.org/10.1007/978-3-662-48971-0_14
Publikationsstatus	Veröffentlicht - 27.11.2015
Veranstaltung	ISAAC 2015 - Nagoya Marriott Associa Hotel , Nagoya, Japan Dauer: 09.12.2015 → 11.12.2015 http://www.al.cm.is.nagoya-u.ac.jp/isaac2015/

UN SDGs

Dieser Output leistet einen Beitrag zu folgendem(n) Ziel(en) für nachhaltige Entwicklung

Zugriff auf Dokument

10.1007/978-3-662-48971-0_14

https://link.springer.com/chapter/10.1007/978-3-662-48971-0_14

Zitieren

@inproceedings{4aab1c7833fd4e1980d8d78b4629df6b,

title = "Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples",

abstract = "Proper learning from positive samples is a basic ingredient for designing secure steganographic systems for unknown covertext channels. In addition, security requirements imply that the hypothesis should not contain false positives. We present such a learner for k-term DNF formulas for the uniform distribution and a generalization to q-bounded distributions. We briefly also describe how these results can be used to design a secure stegosystem.",

author = "Matthias Ernst and Maciej Liskiewicz and R{\"u}diger Reischuk",

year = "2015",

month = nov,

day = "27",

doi = "10.1007/978-3-662-48971-0_14",

language = "English",

isbn = "978-3-662-48970-3",

volume = "9472 ",

series = "Lecture Notes in Computer Science",

publisher = "Springer Verlag",

pages = "151--162",

editor = "Elbassioni, {Khaled } and Makino, {Kazuhisa }",

booktitle = "Algorithms and Computation",

note = "ISAAC 2015 ; Conference date: 09-12-2015 Through 11-12-2015",

url = "http://www.al.cm.is.nagoya-u.ac.jp/isaac2015/",

}

Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples. / Ernst, Matthias; Liskiewicz, Maciej; Reischuk, Rüdiger.
Algorithms and Computation. Hrsg. / Khaled Elbassioni; Kazuhisa Makino. Band 9472 Springer Verlag, 2015. S. 151-162 (Lecture Notes in Computer Science).

Publikation: Kapitel in Büchern/Berichten/Konferenzbänden › Konferenzbeitrag › Begutachtung

TY - GEN

T1 - Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples

AU - Ernst, Matthias

AU - Liskiewicz, Maciej

AU - Reischuk, Rüdiger

PY - 2015/11/27

Y1 - 2015/11/27

N2 - Proper learning from positive samples is a basic ingredient for designing secure steganographic systems for unknown covertext channels. In addition, security requirements imply that the hypothesis should not contain false positives. We present such a learner for k-term DNF formulas for the uniform distribution and a generalization to q-bounded distributions. We briefly also describe how these results can be used to design a secure stegosystem.

AB - Proper learning from positive samples is a basic ingredient for designing secure steganographic systems for unknown covertext channels. In addition, security requirements imply that the hypothesis should not contain false positives. We present such a learner for k-term DNF formulas for the uniform distribution and a generalization to q-bounded distributions. We briefly also describe how these results can be used to design a secure stegosystem.

U2 - 10.1007/978-3-662-48971-0_14

DO - 10.1007/978-3-662-48971-0_14

M3 - Conference contribution

SN - 978-3-662-48970-3

VL - 9472

T3 - Lecture Notes in Computer Science

SP - 151

EP - 162

BT - Algorithms and Computation

A2 - Elbassioni, Khaled

A2 - Makino, Kazuhisa

PB - Springer Verlag

T2 - ISAAC 2015

Y2 - 9 December 2015 through 11 December 2015

ER -

Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples

Abstract

UN SDGs

Zugriff auf Dokument

Fingerprint

Zitieren