Abstract. Many applications of machine learning involve sparse highdimensional data, where the number of input features is (much) larger than the number of data samples, d n. Predi...
Abstract. This paper proposes a general local learning framework to effectively alleviate the complexities of classifier design by means of “divide and conquer” principle and ...
Abstract. We demonstrate that selective attention can improve learning. Considerably fewer samples are needed to learn a source separation problem when the inputs are pre-segmented...
This paper describes a two-stage system for the recognition of postage meter values. A feed-forward Neural Abstraction Pyramid is initialized in an unsupervised manner and trained...
This paper describes a two-stage system for the recognition ge meter values. A feed-forward Neural Abstraction Pyramid is initialized in an unsupervised manner and trained in a sup...