This paper offers a recurrent neural network to support vector machine (SVM) learning in stochastic support vector regression with probabilistic constraints. The SVM is first converted into an equivalent quadratic programming (QP) formulation in linear and nonlinear cases. An artificial neural network for SVM learning is then proposed. The presented neural network framework guarantees obtaining the optimal solution of the SVM problem. The existence and convergence of the trajectories of the network are studied. The Lyapunov stability for the considered neural network is also shown. The efficiency of the proposed method is shown by three illustrative examples.
The heat generation from a notch during the compression-compression fatigue of a cellular Al alloy has been measured and compared
with a model. The measurements indicate that heat is generated because of hysteresis occurring in narrow cyclic plastic zones
outside the notch. This process continues until the notch closes. At closure, a brief period of heat generation arises because
of friction along the notch faces. A plasticity model based on the Dugdale zone is shown to provide a reasonably accurate
characterization of the heat generated, with the proviso that an “ineffective” zone be transposed onto the notch tip. It is
found that the temperatures generated are too small to cause fatigue by thermal softening. A fatigue mechanism based on either
geometric softening of the cells or crack growth in the cell walls is implied. 相似文献
In the current study, we introduce fractional-order Boubaker polynomials related to the Boubaker polynomials to achieve the numerical result for pantograph differential equations of fractional order in any arbitrary interval. The features of these polynomials are exploited to construct the new fractional integration and pantograph operational matrices. Then these matrices and least square approximation method are used to reorganize the problem to a nonlinear equations system which can be resolved by means of the Newton’s iterative method. The brief discussion about errors of the used estimations is deliberated and, finally, some examples are included to demonstrate the validity and applicability of our method.
A synchronized per survivor processing (PSP) maximum likelihood sequence detection (MLSD) receiver is presented for a fast frequency-selective Rayleigh fading channel and large carrier frequency offset. It is shown that, by incorporating the synchronization parameters (the timing error, carrier frequency and phase offsets) into the channel-impulse response (CIR) and modifying the resulting state-space model by a differential approach, the time-varying carrier frequency offset can be effectively tracked using a simple least mean square (LMS) estimator. New criteria are proposed based on the smoothness of the estimated channel derivatives. New criteria are proposed based on this fact that, for smoothly time-varying CIR and frequency offset, the magnitude of the CIR time derivative compared with the magnitude of CIR should be very small. It is illustrated that the new criteria and the conventional mean square error (MSE) are complementary 相似文献
Neural Processing Letters - In this paper, a hybrid scheme based on recurrent neural networks for approximate coefficients (parameters) of ridge fuzzy regression model with LR-fuzzy output and... 相似文献
With the expanding use of the Internet of Things (IoT) devices and the connection of humans and devices to the Internet, the need to provide security in this field is constantly growing. The conventional cryptographic solutions need the IoT device to store secret keys in its non-volatile memory (NVM) leading the system to be vulnerable to physical attacks. In addition, they are not appropriate for IoT applications due to their complex calculations. Thus, physically unclonable functions (PUFs) have been introduced to simultaneously address these issues. PUFs are lightweight and easy-to-access hardware security primitives which employ the unique characteristics of integrated circuits (ICs) to generate secret keys. Among all proposed PUFs, ring oscillator PUF (RO-PUF) has had a more suitable structure for hardware implementation because of its high reliability and easier providing of circuital symmetry. However, RO-PUF has not been so attractive for authentication purposes due to its limited supported challenge-response pairs (CRPs). A few efforts have been made in recent years that could successfully improve the RO-PUF CRP space, such as configurable RO-PUF (CRO-PUF). In this paper, by considerably improving the CRO-PUF structure and adding spare paths, we propose a novel strong RO-PUF structure that exponentially grows the CRP space and dramatically reduces the hardware cost. We implement our design on a simple and low-cost FPGA chip named XC6SLX9-2tqg144, stating that the proposed design can be used in IoT applications. In addition, to improve the CRP space, our design creates a suitable improvement in different security/performance terms of the generated responses, and dramatically outperforms the state-of-the-art. The average reliability, uniqueness, and uniformity of the responses generated are 99.55%, 48.49%, and 50.99%, respectively. 相似文献
Functionally gradient materials are characterized by uniform changes in composition, crystallinity, and/or grain structure,
which may provide unique biological, chemical, or mechanical functionalities in next-generation medical devices. In this article,
the development of functionally gradient Zr−Nb alloys, hydroxyapatite coatings, and diamondlike carbon-metal coatings for
medical applications is reviewed. 相似文献
This study aims to observe the researchers’ behavior in Iranian scientific databases to determine the research gaps and priorities in their field of research. Text mining and natural language processing techniques were used to identify what researchers are looking for and to analyze existing research works. In this paper, the information about the behavior of researchers who work in the field of environmental science and existing research works in the Iranian scientific database are processed. The search trends in all areas are evaluated by analyzing the users’ search data. The trend analysis indicates that in the period of February 2013 to July 2015, the growth of the researchers’ requests in some domains of the environment such as Industry, Training, Assessment, Material, Water and Pollution was 1.5 up to 2 times more than the overall requests. A Combination of the trend analysis and clustering of queries led to shaping four priority zones. Then, the research priorities for each environmental research area were determined. The results show that Training, Pollution, Rangeland, Management and Law are those domains in the environmental research which have the most research gaps in Iran, but there are enough research in Forest, Soil and Industry domains. At the end, we describe the steps for the implementation of a decision support system in environmental research management. Researchers, managers and policy makers can use this proposed “research demand and supply monitoring” system or RDSM to make appropriate decisions and allocate their resources more efficiently. 相似文献