Mutual Information (MI) and Conditional (CMI) are multi-purpose tools from information theory that able to naturally measure the statistical dependencies between random variables, thus they usually of central interest in several machine learning tasks, such as conditional independence testing representation learning. However, estimating CMI, or even MI, is infamously challenging due intractable...

The Scale Invariant Feature Transform (SIFT) operator's success for computer vision applications makes it an attractive solution for the intricate feature based SAR image registration problem. For SAR images, SIFT feature matching results into lot of false alarms. To overcome the mentioned problem, we propose to use mutual information (MI) along with the SIFT operator for SAR image registration...

The classical information-theoretic measures such as the entropy and the mutual information (MI) are widely applicable to many areas in science and engineering. Csiszar generalized the entropy and the MI by using the convex functions. Recently, we proposed the grid occupancy (GO) and the quasientropy (QE) as measures of independence. The QE explicitly includes a convex function in its definitio...

Information theoretic approaches are increasingly being used for reconstructing regulatory networks from microarray data. These approaches start by computing the pairwise mutual information (MI) between all gene pairs. The resulting MI matrix is then manipulated to identify regulatory relationships. A barrier to these approaches is the time-consuming step of computing the MI matrix. We present ...

This paper proposes a method for computing disparity maps from a multimodal stereovision system composed of an infrared and a visible camera pair. The method uses mutual information (MI) as the basic similarity measure where a segmentation-based adaptive windowing mechanism is proposed for greatly enhancing the results. On several datasets, we show that (i) our proposal improves the quality of ...

since the rate of return on different opportunities for investment is random, the optimal distribution of ones capital is put foreward as a stochastic control problem. in this paper, using modem techniques in the theory of random functions and the stochsatic calculus, the problem is discussed for the case when the information possessed by the different investors, as well as their probability be...

Template matching algorithms represent a viable tool to locate particles in optical images. A crucial factor of the performance of these methods is the choice of the similarity measure. Recently, it was shown in [Gao and Helgeson, Opt. Express 22 (2014)] that the correlation coefficient (CC) leads to good results. Here, we introduce the mutual information (MI) as a nonlinear similarity measure ...

Feature selection (FS) is a fundamental task for text classification problems. Text feature aims to represent documents using the most relevant features. This process can reduce size of datasets and improve performance machine learning algorithms. Many researchers have focused on elaborating efficient FS techniques. However, proposed approaches are evaluated small validated single machines. As ...