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.
Word2vec is a novel technique for the study and application of natural language processing(NLP). It trains a word embedding neural network model with a large training corpus. After the model is trained, each word is represented by a vector in the specified vector space. The vectors obtained possess many interesting and useful characteristics that are implicitly embedded with the original words. The...
This paper examines the application of a deep learning approach to converting night-time images to day-time images. In particular, we show that a convolutional neural network enables the simulation of artificial and ambient light on images. In this paper, we illustrate the design of the deep neural network and some preliminary results on a real indoor environment and two virtual environments rendered...
Confidence measures estimate unreliable disparity assignments performed by a stereo matching algorithm and, as recently proved, can be used for several purposes. This paper aims at increasing, by means of a deep network, the effectiveness of state-of-the-art confidence measures exploiting the local consistency assumption. We exhaustively evaluated our proposal on 23 confidence measures, including...
The past few years have seen a dramatic increase in the performance of recognition systems thanks to the introduction of deep networks for representation learning. However, the mathematical reasons for this success remain elusive. A key issue is that the neural network training problem is nonconvex, hence optimization algorithms may not return a global minima. This paper provides sufficient conditions...
Generating high-resolution, photo-realistic images has been a long-standing goal in machine learning. Recently, Nguyen et al. [37] showed one interesting way to synthesize novel images by performing gradient ascent in the latent space of a generator network to maximize the activations of one or multiple neurons in a separate classifier network. In this paper we extend this method by introducing an...
The safety and reliability of roller bearing always have significant importance in rotating machinery. It is needful to build an efficient and excellent accuracy method to monitoring and diagnosis the baring failure. A novel method is presented in this paper to classify the fault feature by wavelet function and extreme learning machine(ELM) that take into account the high accuracy and efficient. The...
Unsupervised learning is a good neural network training way. However, the unsupervised learning algorithm is rare. The generative model is an interesting algorithm which can generate the similar data as the sample data by building a probabilistic model of the input data, and it can be used for unsupervised learning. Variational autoencoder is a typical generative model which is different from common...
This paper proposes an improved method for DBN, by means of introducing the detachment rate. The introduction of detachment rate can play a similar average role, and can make the complex relationship between the neurons weakened, so that DBN learning has stronger robustness. Three kinds of data (corresponding to healthy, faulted and deteriorating) were classified by the improved depth belief network...
One of the key technologies to take full advantage of wind power is to establish a wind turbine (WT) generator output estimation system with high accuracy. The static feed forward artificial neural network is widely used in previous WT generator output estimation technology. However, this method has many problems such as local minimization, a lack of dynamics, edge effect, and multi-correlation. To...
With the development of network technology and e-commerce, online-purchasing has become a fashion which takes a significant ratio of the whole market. Product reviews in e-market platform have a lot of information, and buyers tend to rely on the product'information and the reviews to determine the exactly quality of the product. However, the existence of fake reviews will mislead the consumers and...
This paper addresses the problem of weakly supervised semantic image segmentation. Our goal is to label every pixel in a new image, given only image-level object labels associated with training images. Our problem statement differs from common semantic segmentation, where pixel-wise annotations are typically assumed available in training. We specify a novel deep architecture which fuses three distinct...
Linking two data sources is a basic building block in numerous computer vision problems. Canonical Correlation Analysis (CCA) achieves this by utilizing a linear optimizer in order to maximize the correlation between the two views. Recent work makes use of non-linear models, including deep learning techniques, that optimize the CCA loss in some feature space. In this paper, we introduce a novel, bi-directional...
Japan has the highest debt-to-GDP ratio among advanced countries. Japanese central government debt has increased rapidly in the past 20 years. According to the data provided by Ministry of Finance, Japan, the total central government debt (TCGD) of Japan reached ¥1, 066,423.4 billion (December 31, 2016). This number refreshes the history record of Japanese central government debt. It is important...
Ship detection is a fundamental task for SAR-based maritime surveillance. Besides providing high reliability, a good detector is required to be computationally light, in order to analyze huge areas in a reasonable time. We propose a fully convolutional neural network for ship detection in SAR images. Thanks to a relatively simple architecture, complexity remains low enough to allow for a single-stage...
Convolutional neural networks (CNNs) have shown great potential for remote sensing image classification. As the features obtained from a deep CNN generally exhibit high generalization capacity, the subsequent classifier is normally able to provide good results without the need for careful optimization. However it is well-known that, in the pursuit of high classification results, it is generally difficult...
Understanding temporal expressions is the important foundation of many NLP tasks. However, the varied representations of temporal expressions is difficulty in analysis and understanding. To parsing expressions, an effective classification method of temporal expressions is significant. A temporal expression may belong to one or more classes, but the classification usually requires manual annotation...
The Diversified Sensitivity-based Undersampling (DSUS) is an undersampling method to solve the imbalance pattern classification problems which overcomes the drawbacks of ignoring the distribution information of the training dataset in random-based undersampling methods. The DSUS trains multiple neural networks during the undersampling process. However, only the final one is used. In this work, we...
There are many attempts that utilize deep learning methods to solve the problem of classification in remote sensing images. Convolutional Neural Networks (CNN) have made very good performance for various visual tasks, and marked their important place in all deep learning models. However, for some classification tasks of remote sensing images, CNN could not demonstrate their full potential because...
Hexacopter is a member of rotor-wing Unmanned Aerial Vehicle (UAV) which has 6 six rotors with fixed pitch blades and nonlinear characteristics that cause controlling the attitude of hexacopter is difficult. In this paper, Modified Elman Recurrent Neural Network (MERNN) is used to control attitude and altitude of Heavy-lift Hexacopter to get better performance than Elman Recurrent Neural Network (ERNN)...
Event-Related Potentials (ERPs) is a regular electrophysiological response which is evoked by outer world events or stimuli from brain, and an important approach to explore the human cognitive function. A variety of methods have been proposed for an attempt to analyze it, with varying degrees of success. In this paper, we have proposed a novel method for learning the ERPs, which bases on the effective...
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.