The minimum sum-of-squares clustering problem (MSSC) consists of partitioning n observations into k clusters in order to minimize the sum of squared distances from the points to the centroid of their cluster. In this paper, we propose an exact algorithm for the MSSC problem based on the branch-and-bound technique. The lower bound is computed by using a cutting-plane procedure in which valid inequalities are iteratively added to the Peng–Wei semidefinite programming (SDP) relaxation. The upper bound is computed with the constrained version of k-means in which the initial centroids are extracted from the solution of the SDP relaxation. In the branch-and-bound procedure, we incorporate instance-level must-link and cannot-link constraints to express knowledge about which data points should or should not be grouped together. We manage to reduce the size of the problem at each level, preserving the structure of the SDP problem itself. To the best of our knowledge, the obtained results show that the approach allows us to successfully solve, for the first time, real-world instances up to 4,000 data points.

Oil spills, caused by accidents or by ships cleaning their tanks, represent big threats for maritime and coastal ecosystems health. A very effective detection of oil spills can be performed using satellite synthetic aperture radar (SAR) systems, operating regardless of cloud coverage and sunlight and capable of discriminating oil from regular sea surface. However, discriminating between real oil spills and lookalikes (such as natural oils and seepages, often occurring in upwelling sea areas), although well performed by expert SAR image interpreters, poses a great challenge for automatic processes. In addition, a visual check performed by human operators on a great number of images would be too expensive. Therefore, many solutions for automatic detection have been tried in the last few years, using probabilistic models and, more recently, machine learning. This work presents an innovative solution based on image-to-image translation using convolutional neural networks (CNNs) trained with an adversarial loss function. The proposed approach has been tested, with very promising results, using Radarsat-2 and Sentinel-1 SAR data over the Mediterranean Sea and some areas of the Atlantic Ocean and the North Sea.

Group decisions are a common phenomenon in modern society. Groups typically have increased sensing and cognition capabilities that allow them to make better decisions than individuals. In the last decade, researchers have started evaluating the possibilities to exploit, by means of collaborative BCI, the neurophysiological signals of a group of observers to improve the decision-making process. Following this stream of research, in this work we build a collaborative BCI and focus on the role of each subject in the group, trying to answer the question whether there are some subjects that would be better to remove or that are fundamental to enhance group performance. Moreover, we focus on identifying the smallest subgroup allowing to achieve 100% of accuracy.

In this paper, we consider the feature ranking problem, where, given a set of training instances, the task is to associate a score with the features in order to assess their relevance. Feature ranking is a very important tool for decision support systems, and may be used as an auxiliary step of feature selection to reduce the high dimensionality of real-world data. We focus on regression problems by assuming that the process underlying the generated data can be approximated by a continuous function (for instance, a feedforward neural network). We formally state the notion of relevance of a feature by introducing a minimum zero-norm inversion problem of a neural network, which is a nonsmooth, constrained optimization problem. We employ a concave approximation of the zero-norm function, and we define a smooth, global optimization problem to be solved in order to assess the relevance of the features. We present the new feature ranking method based on the solution of instances of the global optimization problem depending on the available training data. Computational experiments on both artificial and real data sets are performed, and point out that the proposed feature ranking method is a valid alternative to existing methods in terms of effectiveness. The obtained results also show that the method is costly in terms of CPU time, and this may be a limitation in the solution of large-dimensional problems.