scholarly journals Semi-Supervised Training of Transformer and Causal Dilated Convolution Network with Applications to Topic Classification

2021 ◽  
Vol 11 (12) ◽  
pp. 5712
Author(s):  
Jinxiang Zeng ◽  
Du Zhang ◽  
Zhiyi Li ◽  
Xiaolin Li

Aiming at the audio event recognition problem of speech recognition, a decision fusion method based on the Transformer and Causal Dilated Convolutional Network (TCDCN) framework is proposed. This method can adjust the model sound events for a long time and capture the time correlation, and can effectively deal with the sparsity of audio data. At the same time, our dataset comes from audio clips cropped by YouTube. In order to reliably and stably identify audio topics, we extract different features and different loss function calculation methods to find the best model solution. The experimental results from different test models show that the TCDCN model proposed in this paper achieves better recognition results than the classification using neural networks and other fusion methods.

Author(s):  
Denis Voloshinov ◽  
Konstantin Solomonov

The article is devoted to the consideration of a number of issues of hardware and software implementation of constructive geometric models. A rich arsenal of theoretical research in the field of constructive geometry has not been properly used for a long time due to the lack of tools for translating such models using computer technology. The development and improvement of the Simplex geometric modeling system, in which any geometric design is considered as a converter of information represented by signals of a geometric nature, has opened the possibility of applying the achievements of geometric science in computing applications, as well as the development of hardware that implements geometric calculation methods and provides a new graphical interface. The concept developed by the authors is aimed at creating specialized accelerators of geometric transformations.


2019 ◽  
Vol 11 (22) ◽  
pp. 6202 ◽  
Author(s):  
Valentina Zaccaria ◽  
Moksadur Rahman ◽  
Ioanna Aslanidou ◽  
Konstantinos Kyprianidis

The correct and early detection of incipient faults or severe degradation phenomena in gas turbine systems is essential for safe and cost-effective operations. A multitude of monitoring and diagnostic systems were developed and tested in the last few decades. The current computational capability of modern digital systems was exploited for both accurate physics-based methods and artificial intelligence or machine learning methods. However, progress is rather limited and none of the methods explored so far seem to be superior to others. One solution to enhance diagnostic systems exploiting the advantages of various techniques is to fuse the information coming from different tools, for example, through statistical methods. Information fusion techniques such as Bayesian networks, fuzzy logic, or probabilistic neural networks can be used to implement a decision support system. This paper presents a comprehensive review of information and decision fusion methods applied to gas turbine diagnostics and the use of probabilistic reasoning to enhance diagnostic accuracy. The different solutions presented in the literature are compared, and major challenges for practical implementation on an industrial gas turbine are discussed. Detecting and isolating faults in a system is a complex problem with many uncertainties, including the integrity of available information. The capability of different information fusion techniques to deal with uncertainty are also compared and discussed. Based on the lessons learned, new perspectives for diagnostics and a decision support system are proposed.


Author(s):  
Hans-Joachim Winkel ◽  
Mathias Paschen

Modern nets consist of meshes made of threads or twines with spirals or helical strakes. Fluid-structure interactions have been investigated in Rostock for a long time applying different theoretical models. Because of great net flexibility there is a need of calculation methods which consider the main physical qualities. This is done by the approximation of wake of threads by results from circular cylinders and influence of circulation, which is known from measurements of transverse force. Results of measurements with two models with and without spirals are given for comparison.


2019 ◽  
Vol 27 (4) ◽  
pp. 1103-1114
Author(s):  
Wendy Marsh ◽  
Ann Robinson ◽  
Jill Shawe ◽  
Ann Gallagher

Background Midwives and nurses appear vulnerable to moral distress when caring for women whose babies are removed at birth. They may experience professional dissatisfaction and their relationships with women, families and colleagues may be compromised. The impact of moral distress may manifest as anger, guilt, frustration, anxiety and a desire to give up their profession. While there has been much attention exploring the concept of moral distress in midwifery, this is the first study to explore its association in this context. Aim This article explores midwives’ experiences of moral distress when providing care to women whose babies were removed at birth and gives valuable insight into an issue nurses and midwives encounter in their profession. Methods Four mothers and eight midwives took part in this research. Narrative inquiry incorporating photo-elicitation techniques was used to generate data; mothers were interviewed face to face and midwives through focus groups. The images and audio data were collected, transcribed and analysed for emerging themes. For the purpose of this article, only the midwives’ stories are reported. This research received a favourable ethical opinion from the University of Surrey Ethics committee. Ethical considerations This study received a favourable ethical approval from a higher education institutes ethics committee. Results Midwives who care for women whose babies are removed at birth report it as one of the most distressing areas of contemporary clinical practice. Furthermore, they report feelings of guilt, helplessness and betrayal of the midwife–mother relationship. Many of the midwives in this study state that these experiences stay with them for a long time, far more than more joyful aspects of their role. Conclusion Midwives experience moral distress. Support systems, education and training must be available to them if we are to reduce the long-term impact upon them, alleviate their distress and prevent them from leaving the profession.


2000 ◽  
Vol 12 (07) ◽  
pp. 921-944 ◽  
Author(s):  
JOHAN ANDRIES ◽  
FABIO BENATTI ◽  
MIEKE De COCK ◽  
MARK FANNES

In this paper, we consider the long time asymptotics of multi-time correlation functions for quantum dynamical systems that are sufficiently random to relax to a reference state. In particular, the evolution of such systems must have a continuous spectrum. Special attention is paid to general dynamical clustering conditions and their consequences for the structure of fluctuations of temporal averages. This approach is applied to the so-called Powers–Price shifts.


Sign in / Sign up

Export Citation Format

Share Document