Extraction of attributes for visual object recognition and DNA microarray analysis

We introduce a new model formalizing selection of features from a large dictionary of variables that can be computed from a signal or an image. Features are extracted according to an efficiency criterion, on the basis of specified classification tasks. We estimate a probability distribution P on the complete dictionary, which distributes its mass over the more efficient or informative components. The method is then tested on several problems of signal processing like face detection, handwritten digit recognition and DNA microarray analysis