We examine whether steganographic images can be detected more reliably when there exist other images, taken with the same camera under the same conditions, of the same scene. We argue that such a circumstance is realistic and likely in practice. In ‘laboratory conditions’ mimicking circumstances favourable to the analyst, and with a custom set of digital images which capture the same scenes wit...

Clustering problems often involve datasets where only a part of the data is relevant to the problem, e.g., in microarray data analysis only a subset of the genes show cohesive expressions within a subset of the conditions/features. On such datasets, in order to accurately identify meaningful clusters, both non-informative data points and non-discriminative features need to be discarded. Additio...

Steganalysis aims to detect the presence of hidden information, known as steganography, in digital images. Modern detectors rely on machine learning techniques, using features from digital images to classify them as covers or stegos. Often the ability to detect steganography can be improved through a process called calibration. Calibration attempts to establish a reference for cover features in...

We study families of possibly overlapping self-affine sets. Our main example is a family that can be considered the self-affine version of Bernoulli convolutions and was studied, in the non-overlapping case, by F. Przytycki and M. Urbański [23]. We extend their results to the overlapping region and also consider some extensions and generalizations.

Once a decomposition of the nite element space V into two or more subspaces is given, e.g. , via domain decomposition, a speciic Multiplicative Schwarz Method (MSM) and Additve Schwarz Method (ASM) is deened. In this paper, we analyse the MSM for the decomposition induced by the approximate discrete harmonic nite element basis which was introduced in a joint paper of the authors with A. Meyer (...

The group Lasso is an extension of the Lasso for feature selection on (predefined) non-overlapping groups of features. The non-overlapping group structure limits its applicability in practice. There have been several recent attempts to study a more general formulation, where groups of features are given, potentially with overlaps between the groups. The resulting optimization is, however, much ...

In the past few years co-clustering has emerged as an important data mining tool for two way data analysis. Coclustering is more advantageous over traditional one dimensional clustering in many ways such as, ability to find highly correlated sub-groups of rows and columns. However, one of the overlooked benefits of co-clustering is that, it can be used to extract meaningful knowledge for variou...

We consider the problem of partitioning a graph G = (V; E) into two sets V 1 and V 2 such that jV 1 \ V 2 j is no more than an integer d, and such that P u2V 1 ?V 2 ;v2V 2 ?V 1 (u; v) is minimized. We show that this problem is NP-hard in general. It remains NP-hard if jV 1 j = jV 2 j, or if we insist that, for any bipartition, v 1 2 V 1 and v 2 2 V 2 for two speciic nodes v 1 ; v 2 2 V. The pro...