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

Matthias Ernst, Maciej Liskiewicz, Rüdiger Reischuk

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.

Original languageEnglish
Title of host publicationAlgorithms and Computation
EditorsKhaled Elbassioni, Kazuhisa Makino
Number of pages12
Volume9472
PublisherSpringer Verlag
Publication date27.11.2015
Pages151-162
ISBN (Print)978-3-662-48970-3
ISBN (Electronic)978-3-662-48971-0
DOIs
Publication statusPublished - 27.11.2015
EventISAAC 2015 - Nagoya Marriott Associa Hotel , Nagoya, Japan
Duration: 09.12.201511.12.2015
http://www.al.cm.is.nagoya-u.ac.jp/isaac2015/

Fingerprint

Dive into the research topics of 'Algorithmic Learning for Steganography: Proper Learning of k-term DNF Formulas from Positive Samples'. Together they form a unique fingerprint.

Cite this