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.
Efforts to open and share research outcomes (publications, data, etc.) by public funds have been performed. And efforts to spread knowledge more easily and quickly have been also promoted. However, in Korea, we do not manage the research outcomes as core assets that can be produced and reused by public funds. In this paper, we analyze the characteristics of overseas studies in terms of deposition,...
Face sketch synthesis plays an important role in both law enforcement and digital entertainment. The existing methods for sketch synthesis always suffer from noising and blurring effect. To resolve these problems, a nonsubsampled Shearlet transform (NSST) based detail enhancement strategy is proposed. The exemplar-based method is firstly adopted to synthesize the primary sketch, then the final sketch...
In this paper, we try to recognize negative emotions (sadness and disgust) of human affecting driving by using physiological signals that are commonly used to deal with human emotions. To do this, emotional stimuli are used to induce sadness and disgust, and emotion recognition is performed based on the feature vector extracted from the physiological signals collected on the induced emotion by a stimulus...
In this paper, we attempt to manage GC overhead at the operating system level. In our approach, first, we use a machine learning technique to devise a GC detecting mechanism at the operating system level, and second, we show that by making use of this mechanism performance variance normally observed on SSDs can be reduced. We develop a GC-detector that detects garbage collection of SSDs and request...
Machine learning is currently a hot research topic and applied in intelligence transportation system to discover new valuable knowledge and patterns. In this paper, we extract trajectory information from popular traffic simulator and apply it into four different machine learning methods. In the case of the Gangnam district in Seoul, the Gradient Boosting Regression has better fit with lower values...
Modern control theories such as systems engineering approaches try to solve nonlinear system problems by revelation of causal relationship or co-relationship among the components; most of those approaches focus on control of sophisticatedly modeled white-boxed systems. We suggest an application of actor-critic reinforcement learning approach to control a nonlinear, complex and black-boxed system....
This paper presents the multi-site photovoltaic (PV) power generation forecast using the deep-learning algorithms. South Korea PV power generation is the most popular among renewable energy sources as government policy. Form of PV power business is small scale which need to forecast the generation of energy. This paper present the forecast model for multi-site PV power plant using deep-learning algorithm...
Recently many industries and companies are developing machine learning algorithms and services, and they are publishing them on the internet. However, because most of people who want to use the machine learning services to analyze data are familiar with sheet data rather than programming language, it is difficult to use those services written in programming language. For the reason, we developed a...
In this paper, we introduce seven emotions and positive and negative emotion recognition methods using facial images and the development of apps based on the method. In previous researches, they used the deep-learning technology to generate models with emotion-based facial expressions to recognized emotions. There are existing apps that express six emotions, but not seven emotions and positive and...
We propose a gray coding method for deep neural network (DNN) based decoder. With multiple resources considered together, DNN can be used to decode corrupted signals. In deep learning training, stochastic gradient descent (SGD) algorithm is used, which means that the cost function must be differentiable. Then, allocating the discrete bits for each symbol is difficult. To solve this problem, the basic...
Within the world of wireless technologies, Bluetooth has recently been at the forefront of innovation. It is becoming increasingly relevant for vehicles to become aware of their surroundings. Therefore, having knowledge of nearby Bluetooth devices, both inside and outside other vehicles, can provide the listening vehicles with enough data to learn about their environment. In this paper, we collect...
Recently, various technologies related to the 4th Industrial Revolution (cloud, Big Data, Internet of Things, artificial intelligence, etc.) have become issues and deep learning has become a favorite technique for big data and the studies using related techniques have been conducted on astronomy, physics, Science, and statistical analysis. The literature published by the researchers is increasingly...
In this paper, we propose a classification model for learning state based on individual biometric data. In particular, we use the pupil size as a biometric data and the data has been collected from 72 participants. We also deploy the support vector machine (SVM) in conjunction with k-fold validation as an analysis tool. In order to improve the performance of the SVM, the we remove outliers from the...
In distributed DNN training, the speed of reading and updating model parameters greatly affects model training time. In this paper we investigate the performance of deep neural network training with parameter sharing based on shared memory for distributed machine learning. We propose a shared memory-based modification of the deep learning framework. In our framework, remote shared memory is used to...
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.