Counterfactual visual explanations are aimed at identifying changes in an image that will modify the prediction of a classifier.Unlike adversarial images, counterfactuals are required to be realistic.For this reason generative models such as variational autoencoders (VAE) have been used to restrain the search of counterfactuals on the data manifold.However such gradient-based approaches remain limited even when they deal with simple datasets such as MNIST.Conjecturing that these limitations result from a plateau effect which makes the gradient noisy and less informative, we improve the gradient estimation by training an ensemble of classifiers directly in the latent space of VAEs.Several experiments show that the resulting method called Clarity delivers counterfactual images of high-quality, competitive with the state-of-the-art.

We consider the task of discovering the top-K reliable approximate functional dependencies X -> Y from high dimensional data. While naively maximizing mutual information involving high dimensional entropies over empirical data is subject to false discoveries, correcting the empirical estimator against data sparsity can lead to efficient exact algorithms for robust dependency discovery. Previous approaches focused on correcting by subtracting expected values of different null hypothesis models. In this paper, we consider a different correction strategy and counter data sparsity using uniform priors and smoothing techniques, that leads to an efficient and robust estimating process. In addition, we derive an admissible and tight bounding function for the smoothed estimator that allows us to efficiently solve via branch-and-bound the hard search problem for the top-K dependencies. Our experiments show that our approach is much faster than previous proposals, and leads to the discovery of sparse and informative functional dependencies.

The article proposes a generic method to classify vertices or edges of a labeled graph. More precisely the method computes a confidence index for each vertex v or edge e to be a member of a target class by mining the topological environments of v or e. The method contributes to knowledge discovery since it exhibits for each edge or vertex an informative environnement that explains the found confidence. When applied to the problem of discovering strategic bonds in molecules, the method correctly classifies most of the bonds while providing relevant explanations to chemists. The developed algorithm GemsBond outperforms both speed and scalability of the learning method that has previously been applied to the same application while giving similar results.

This paper describes a software architecture for real-world robotic applications. We discuss issues of software reliability, testing and realistic off-line simulation that allows the majority of the automation system to be tested off-line in the laboratory before deployment in the field. A recent project, the automation of a very large mining machine is used to illustrate the discussion.

The mining industry is highly suitable for the application of robotics and automation technology since the work is arduous, dangerous and often repetitive. This paper discusses a robust sensing system developed to find and trade the position of the hoist ropes of a dragline. Draglines are large 'walking cranes' used in open-pit coal mining to remove the material covering the coal seam. The rope sensing system developed uses two time-of-flight laser scanners. The finding algorithm uses a novel data association and tracking strategy based on pairing rope data.