Learn More
We present an approach to feature extraction that is a generalization of the classical linear discriminant analysis (LDA) on the basis of deep neural networks (DNNs). As for LDA, discriminative features generated from independent Gaussian class conditionals are assumed. This modeling has the advantages that the intrinsic dimensionality of the feature space(More)
Deep Neural Networks (DNNs) denote multilayer artificial neural networks with more than one hidden layer and millions of free parameters. We propose a Generalized Discriminant Analysis (GerDA) based on DNNs to learn discriminative features of low dimension optimized with respect to a fast classification from a large set of acoustic features for emotion(More)
Drosophila polytene interphase chromosomes provide an ideal test system to study the rules that define the structure of chromatin domains. We established a transgenic condensed chromatin domain cassette for the insertion of large pieces of DNA by site-specific recombination. Insertion of this cassette into open chromatin generated a condensed domain,(More)
Eukaryotic chromatin is organized in contiguous domains that differ in protein binding, histone modifications, transcriptional activity, and in their degree of compaction. Genome-wide comparisons suggest that, overall, the chromatin organization is similar in different cells within an organism. Here, we compare the structure and activity of the 61C7-61C8(More)
(d) (e) Fig. 11 Pedestrian tracking (a) (b) (c) (d) (e) Fig. 12 Vehicle tracking achieves superior speed and accuracy through the introduction of several performance enhancing techniques. The method-specific optimizations also enhance overall system performance without affecting worst-case execution times. These optimizations apply to various region-based(More)