Automatic extraction of distinctive features from a visual information stream is challenging due to the large amount of information contained in most image data. In recent years deep neural networks (DNNs) have gained outstanding popularity for solving visual information processing tasks. This study reports novel contributions, including a new DNN architecture and training method, which increas...

Recent studies have shown that deep neural networks (DNNs) perform significantly better than shallow networks and Gaussian mixture models (GMMs) on large vocabulary speech recognition tasks. In this paper, we argue that the improved accuracy achieved by the DNNs is the result of their ability to extract discriminative internal representations that are robust to the many sources of variability i...

Raw temporal features were derived from extracted temporal envelope bank (referred to as “Tbank”). Tbank features were used with deep neural networks (DNNs) to greatly increase the amount of detailed information about the past to be carried forward to help in the interpretation of the future.

A method based on the operation of so called dynamic neural networks (DNN) for music recommendation is described. DNNs are trained with the signals of each melody and not with traditional descriptors. The method has been tested with a database composed of 1.200 melodies, at different sampling frequencies.

This paper is concerned with the stability problem with two additive time-varying delay components. By choosing one augmented Lyapunov-Krasovskii functional, using some new zero equalities, and combining linear matrix inequalities (LMI) techniques, two new sufficient criteria ensuring the global stability asymptotic stability of DNNs is obtained. These stability criteria are present in terms of...

INTrODUCTION Recurrent neural networks whose neurons are fully interconnected have been utilized to implement associative memories and solve optimization problems. These networks are regarded as nonlinear dynamical feedback systems. Stability properties of this class of dynamical networks are an important issue from applications point of view. ABSTrACT Global stability analysis for complex-valu...

Recently, several optimization methods have been successfully applied to the hyperparameter optimization of deep neural networks (DNNs). The methods work by modeling the joint distribution of hyperparameter values and corresponding error. Those methods become less practical when applied to modern DNNs whose training may take a few days and thus one cannot collect sufficient observations to accu...

Speech emotion recognition is a challenging problem partly because it is unclear what features are effective for the task. In this paper we propose to utilize deep neural networks (DNNs) to extract high level features from raw data and show that they are effective for speech emotion recognition. We first produce an emotion state probability distribution for each speech segment using DNNs. We th...

When deployed in automated speech recognition (ASR), deep neural networks (DNNs) can be treated as a complex feature extractor plus a simple linear classifier. Previous work has investigated the utility of multilingual DNNs acting as language-universal feature extractors (LUFEs). In this paper, we explore different strategies to further improve LUFEs. First, we replace the standard sigmoid nonl...

In deep neural networks with convolutional layers, each layer typically has fixed-size/single-resolution receptive field (RF). Convolutional layers with a large RF capture global information from the input features, while layers with small RF size capture local details with high resolution from the input features. In this work, we introduce novel deep multi-resolution fully convolutional neural...