Interpretability in Machine Learning – We present in this paper a statistical procedure that gives the maximum accuracy on the posterior of all the possible outputs of a given model with a fixed amount of data. The procedure is illustrated using a standard dataset, namely the dataset generated with a model with a certain number of parameters. The procedure is illustrated with a model with certain number of parameters.

Most current methods treat a set of discrete observations (e.g., a model and a test) as a collection of observations. Such approaches typically assume that samples are modeled as discrete samples, which may not be the case. In this work we present a new approach for classification experiments based on Bayesian networks, where the classifier is a single distribution over observations. In addition, we present a generalization error measure that enables us to compare the resulting classifiers to a subset of the observed distributions. To the best of our knowledge, our contribution is the first one to analyze data in this manner, outperforming a state-of-the-art classification algorithm in this task.

Learning to See and Feel the Difference

Segmentation and Optimization Approaches For Ensembled Particle Swarm Optimization

Interpretability in Machine Learning

Predicting visual stimuli based on saliency maps

Structural Correspondence Analysis for Semi-supervised LearningMost current methods treat a set of discrete observations (e.g., a model and a test) as a collection of observations. Such approaches typically assume that samples are modeled as discrete samples, which may not be the case. In this work we present a new approach for classification experiments based on Bayesian networks, where the classifier is a single distribution over observations. In addition, we present a generalization error measure that enables us to compare the resulting classifiers to a subset of the observed distributions. To the best of our knowledge, our contribution is the first one to analyze data in this manner, outperforming a state-of-the-art classification algorithm in this task.