In the typical supervised learning scenario we are given a set of labeled examples and we aim to induce a model that captures the regularity between the input and the class. However, most of the classification algorithms require hundreds or even thousands of labeled examples to achieve satisfactory performance. Data labels come at high costs as they require expert knowledge, while unlabeled dat...

A new effective and fast minimization approach completely based on the physical theory is proposed for protein design. The sequence space is essentially searched according to the Boltzmann distribution. In this approach, the relative entropy is used as a minimization object function. The method has been tested on an off-lattice model of proteins and the results are better than those obtained fr...

We analyze proximal methods based on entropy-like distances for the minimization of convex functions subject to nonnegativity constraints. We prove global convergence results for the methods with approximate minimization steps and an ergodic convergence result for the case of finding a zero of a maximal monotone operator. We also consider linearly constrained convex problems and establish a qua...

Entropy-based cost functions are enjoying a growing attractiveness in unsupervised and supervised classification tasks. Better performances in terms both of error rate and speed of convergence have been reported. In this letter, we study the principle of error entropy minimization (EEM) from a theoretical point of view. We use Shannon's entropy and study univariate data splitting in two-class p...

An effective and fast minimization approach is proposed for the prediction of protein folding, in which the 'relative entropy' is used as a minimization function and the off-lattice model is used. In this approach, we only use the information of distances between the consecutive Calpha atoms along the peptide chain and a generalized form of the contact potential for 20 types of amino acids. Tes...