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.
In this paper, balanced two-stage residual networks (BTSRN) are proposed for single image super-resolution. The deep residual design with constrained depth achieves the optimal balance between the accuracy and the speed for super-resolving images. The experiments show that the balanced two-stage structure, together with our lightweight two-layer PConv residual block design, achieves very promising...
We present an anomaly detection system based on an autonomous robot performing a patrol task. Using a generative adversarial network (GAN), we compare the robot's current view with a learned model of normality. Our preliminary experimental results show that the approach is well suited for anomaly detection, providing efficient results with a low false positive rate.
Individual surgeons rely on residency programs as their main conduit for developing the necessary soft skills needed to succeed and excel in the operating room. One critical skill requiring subjective (qualitative) learning involves navigating through varying soft and hard tissues by hand, and, most importantly, understanding how medical instruments respond under these conditions. During residency,...
The convolutional neural network (CNN) is more and more popular in computer vision and widely used in acoustic signal processing, image classification, and image segmentation. In this work, an architecture which is a combination of the 3-D convolutional neural network and the long short term memory (LSTM) was proposed for action recognition. It stacks the consecutive video frames, extracts spatial...
In this paper, we use a advanced method called Faster R-CNN to detect traffic signs. This new method represents the highest level in object recognition, which don't need to extract image feature manually anymore and can segment image to get candidate region proposals automatically. Our experiment is based on a traffic sign detection competition in 2016 by CCF and UISEE company. The mAP(mean average...
The learning of Physics involves building up and using lab experiments. In turn, teachers must be trained in experimenting and using several resources that enable them to design valuable teaching strategies and learning activities. Thanks to Information and Communication Technologies (ICT), virtual and remote labs can provide a framework where physical experiments can be developed. Altough remote...
We study learning outcome prediction for online courses. Whereas prior work has focused on semester-long courses with frequent student assessments, we focus on short-courses that have single outcomes assigned by instructors at the end. The lack of performance data makes the behavior of learners, captured as they interact with course content and with one another in Social Learning Networks (SLN), essential...
Learning distributed representations of symbolic data were introduced by Hinton[1], and first developed in modeling networks for learning the node vectors by Perozzi et al (2014). In this work, we proposed Dnps, a novel nodes embedding approach for acquiring distributed representations of large-scale dynamic social networks. Dnps is suitable for many types of social networks: dynamic/static, directed/undirected,...
It is shown that virtual laboratory complexes identical to real physical stands may provide an alternative in laboratory workshop for students of technical specialties of higher educational establishments. The basic requirements for virtual complexes are formulated and graphic programming medium LabVIEW is proposed as a platform for their creation. An example of a virtual laboratory complex for the...
The paper encourages the development of energy-saving technologies and the improvement of environmental safety of road transport. Expected results: the joint educational program for master's degree of project related speciality has been created and is ready for the implementation; curricular/syllabi, training courses, methodical and teaching materials have been developed; retraining and advanced training...
Educational systems of IT-companies typically provide a network space in which students and teachers can come together for teaching and learning. Using the microservice architecture offers advantages in application development.
This paper shows a simple approach for fake news detection using naive Bayes classifier. This approach was implemented as a software system and tested against a data set of Facebook news posts. We achieved classification accuracy of approximately 74% on the test set which is a decent result considering the relative simplicity of the model. This results may be improved in several ways, that are described...
Web applications commonly provide a high attack surface. In today's world of high impact attacks, protecting them against both known and unknown attacks becomes more important than ever. We present an approach of machine learning based anomaly detection to flexibly detect anomalous requests. Our approach leverages long short-term memory (LSTM) neural networks to learn a detailed model of normal requests...
Improving the accuracy, reducing the time to authenticate users and preserving privacy are some of the pivotal issues in smartphone security. A majority of published owner identification methods have concentrated on improving accuracy, emphasizing less on response time. Usage pattern of smartphone apps by the owner may be used as an important signature to differentiate between the legitimate user...
This paper presents two different implementations for recognition of handwritten numerals using a high performance autoencoder and Principal Component Analysis (PCA) by making use of neural networks. Different from other approaches, the non-linear mapping capability of neural networks is used extensively here. The implementation involves the deployment of a neural network, and the use of an auto encoder...
For i-vector model, normalization approach is Probabilistic linear discriminant analysis and has a significant performance for verification of speaker. However it requires a huge development data which cost a lot in many cases. Unsupervised adaption method is a possible approach, which use unlabeled data to adapt PLDA scattering matrices to the target domain. In this paper, ‘local training’ approach...
Young teachers in colleges and universities usually play important role as main force of college teaching, as well as the mentor of young students. But not only that, they also act as the participants and transferees of social civilization. Under the new situation, the realization of the young-teacher-career-development system would become an important issue in the development of higher education...
After the development of about half a century, Taiwan's industrial design education has entered the international frontier now. This paper summarizes the characteristics of industrial design education in Taiwan universities from the aspects of education scale and department, discipline and professional construction, production and research cooperation, and so on and expects to gain experience and...
To meet the global tourism market trend and educational need, English for Tourism Purposes (ETP) courses have been rapidly developed in Taiwan's universities for recent years. Investigation and evaluation of national-wide ETP curriculum, give directions to the educators whose learners are second language speakers of English and aspire to work in the international tourism and hospitality industry have...
In recent years, Wi-Fi based indoor localization using received signal strength (RSS) gets considerable attention. However, RSS based Wi-Fi localization at 2.4GHz is highly susceptible and unstable. We proposed dynamic machine learning approach (DMLA) at 5GHz to effectively localize Wi-Fi users by means of feed-forward neural network algorithm. The simulation result shows that 90% of tested result...
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.