The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
Restricted Boltzmann Machine (RBM) is the building block of Deep Belief Nets and other deep learning tools. Fast learning and prediction are both essential for practical usage of RBM-based machine learning techniques. This paper presents a concept named generalized redundancy elimination to avoid most of the the computations required in RBM learning and prediction without changing the results. It...
Deep Neural Networks (DNNs) have emerged as a core tool for machine learning. The computations performed during DNN training and inference are dominated by operations on the weight matrices describing the DNN. As DNNs incorporate more stages and more nodes per stage, these weight matrices may be required to be sparse because of memory limitations. The GraphBLAS.org math library standard was developed...
The paper proposes using a neuro-fuzzy controller in telecommunication networks for improving the routing process. An architecture of the neuro-fuzzy controller was developed. Linguistic variables, terms and membership functions for input and output values were defined. A rules base was developed. The operation of the neuro-fuzzy controller was simulated and trained.
Since short text is characterized of the short length, sparse features and strong context dependency, the traditional models have a limited precision. Motivated by this, this article offers an empirical exploration on a character-level model which implements a combination of convolutional neural network(CNN) and recurrent neural networks(RNN) for short text classification. Including the highway networks...
Obtaining ultrafast images using steered plane wave (PW) imaging remains a challenge due to the trade-off between image quality and frame rate. PW imaging indeed relies on compounding in order to preserve a good image quality, usually using multiple successive emissions, which in turn yields a decrease of the frame rate. As opposed to this classical approach, we propose a new strategy to reduce the...
For deep learning applications, large numbers of samples are essential. If this condition is not met, effective features cannot be generated and overfitting occurs especially for the small datasets such as in medical applications. To address this issue, we propose a new dynamic ensemble merging algorithm that iteratively adjusts the weights of a convolutional neural network (CNN) ensemble's elements...
Echocardiograms are acquired from standard views to ensure correct assessment of cardiac function. There is an increasing use of quantitative tools where specific views are required. Further, non-expert users of echocardiography are increasing, and thus a need for quality assurance during imaging. The aim of this project is to develop automatic and robust real-time classification of cardiac views...
Most of the existing algorithm of objective image quality evaluation are often for a single type of distortion, and the effect of multi-distortion image quality evaluation is poor, In this paper presents a no reference image quality assessment method based on phase congruency and convolution neural network, to evaluate the mixed distorted image, Firstly, the input image is divided into blocks and...
This paper presents the new face verification algorithm based on deep convolutional neural network. The algorithm produces face feature vectors, distance between these vectors allows to determine whether images from the same class. Comparative experimental results are given for LFW test database and modern face recognition algorithms. ROC-curve and equal error rate are used to determine the accuracy...
This paper presents a knee torque estimation in non-pathological gait cycle at stance phase. Comparative modelling by using dynamics model and neural network model is discussed. Dynamics modelling is constructed by using simple two degree of freedom dynamics with Newtonian calculation approach and more complex four degree of freedom dynamics with Lagrangian calculation approach. Neural network based...
Convolutional neural network has made major progress in classification problems of general object recognition. Classification of facial images is one of them. However, expression of the networks for classification depends on datasets and the network model, which is vulnerable to changes in the adaptation range. We propose the network that has the two convolutional parts of pre-trained CNN by transfer...
We propose a neural network architecture for depth map inference from monocular stabilized videos with application to UAV videos in rigid scenes. Training is based on a novel synthetic dataset for navigation that mimics aerial footage from gimbal stabilized monocular camera in rigid scenes. Based on this network, we propose a multi-range architecture for unconstrained UAV flight, leveraging flight...
This paper aims at construction of a system which assumes food textures. The system consists of equipment for obtaining the load and the sound signals while the probe is stabbing the food, and the neural network model infers the degree of the food texture. In the experiment, the validity of our proposed system is discussed.
Automatic segmentation of the left ventricle (LV) can become a useful tool in echocardiography. Deep convolutional neural networks (CNNs) have shown promising results for image classification and segmentation on several domains, however CNNs seem to require a lot of training data. In this work, CNNs are investigated for LV ultrasound image segmentation. We study if the need for manual annotation can...
Word2Vec is a popular set of machine learning algorithms that use a neural network to generate dense vector representations of words. These vectors have proven to be useful in a variety of machine learning tasks. In this work, we propose new methods to increase the speed of the Word2Vec skip gram with hierarchical softmax architecture on multi-core shared memory CPU systems, and on modern NVIDIA GPUs...
In this work, we conduct a systematic exploration on the promise and challenges of deep learning for the sparse matrix format selection. We propose a set of novel techniques to solve special challenges to deep learning, including input matrix representations, a late-merging deep neural network structure design, and the use of transfer learning to alleviate cross-architecture portability issues.
Convolutive Non-Negative Matrix Factorization model factorizes a given audio spectrogram using frequency templates with a temporal dimension. In this paper, we present a convolutional auto-encoder model that acts as a neural network alternative to convolutive NMF. Using the modeling flexibility granted by neural networks, we also explore the idea of using a Recurrent Neural Network in the encoder...
Neural networks have demonstrated promising results for a wide range of applications. The proposed techniques employ different architectures and objective functions to adapt to the application while enabling a feasible implementation. Commonly used objective functions for network optimization are based on the cross entropy between the empirical distribution of the training data and the model distribution...
Aiming at the problem that the micro drills is easy to be broken in the process of drilling; it is difficult to detect the drill bit. The drilling torque signal is taken as the monitoring object. A new method for the on-line monitoring the micro-drill breakage based on BP neural network is proposed. After the three layer wavelet decomposition of the drilling torque signal, the energy feature vector...
Extreme Learning Machine (ELM) is a neural network architecture with Single Layer Feed-forward Neural Network (SLFN). For meaningful results, the structure of ELM has to be optimized through the inclusion of regularization and the ℓ2 — norm based regularization is mostly used. ℓ2-norm based regularization achieves better performance than the traditional ELM. The estimate of the regularization parameter...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.