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.
A parameter learning algorithm based on noise data smoothing is developed in static Bayesian Networks (BN) to tackle the problem of randomly missing observed information, i.e., data missing can occur arbitrarily in every group of data in the sample. the simulation results demonstrate that this algorithm has similar speed and accuracy compared with EM algorithm in the condition of missing proportion...
This paper introduces a new algorithm for passivity enforcement of linear lumped macromodels in scattering form. As typical in most state of the art passivity enforcement methods, we start with an initial non-passive macromodel obtained by a Vector Fitting process, and we perturb its parameters to make it passive. The proposed scheme is based on a convex formulation of both passivity constraints and...
Considering the Particle Swam Optimization (PSO) is easily relapsing into local extremum, an improved PSO(IPSO) is proposed in this paper. In the new algorithm, we apply the evolution speed factor as the trigger conditions to stochastically disturb the local optimal solution. The IPSO algorithm can not only improve extraordinarily the convergence velocity in the evolutionary optimization, but also...
According to the common phenomenon that accuracy of range-based location algorithm for WSN could not satisfy the requirement of location accuracy, PSO algorithm is introduced into localization for WSN. Meantime, to solve the premature convergence problem of PSO, improved algorithms with hybrid and mutation operators are proposed, leading to obtain a high level of particle population diversity, decrease...
Markov Chain Monte Carlo (MCMC) has been essential in tracking vehicle undergoing disturbances for traffic surveillance purposes. It is capable of tracking vehicle by estimating the vehicle's position with the sampling of probability distributions. However the accuracy of the position estimation is highly dependent on the sampling efficiency of MCMC. Therefore the sample size of the MCMC is adapted...
Accurate location of target nodes is highly desirable in a Wireless Sensor Network (WSN) as it has a strong impact on overall performance of the WSN. This paper proposes the application of H-Best Particle Swarm Optimization (HPSO) and Biogeography Based Optimization (BBO) algorithms for distributed optimal localization of randomly deployed sensors. The proposed HPSO algorithm is modeled for fast and...
The traditional BP neural network training method processes the training dataset serially on one machine, so the efficiency is quite low. The massive data that need to be explored brings great challenge for BP neural network. The traditional serial training method of BP neural network will encounter many problems, such as costing too much time and insufficient memory to finish the training process...
This paper presents a fast-convergence and robust adaptive step size equalization approach for a 14-bit 200MS/s hybrid pipeline-SAR analog-to-digital converters (ADC). The proposed calibration approach corrects errors not only from capacitor mismatch, gain error, op amp nonlinearity, and comparator offset, but also the reference DAC error and inter-stage mismatch errors. It is robust and permits higher...
In order to realize automatically classifying can defects and improve the convergence speed and the classification accuracy of Self-Organizing Feature Map (SOFM) neural network, 5 improved measures are presented in this paper. They include using typical sample vector, introducing frequency sensitive factor, learning rate adaptive adjustment, selecting convergence criterion and searching winning neuron...
This paper presents three accurate methods for finding simple roots of polynomials in floating point arithmetic. We present them by using the Compensated Horner algorithm to accurately compute the residual which can yield a full precision when the problem is ill-conditioned enough. Some numerical experiments are conducted to justify the proposed approaches.
The conventional algorithm of the BP neural network has some disadvantages such as in the vicinity of the target, if the learning factor is too small, the convergence may be too slow, and if the learning factor is too large, the convergence may be amended too much, leading to oscillations and even dispersing phenomenon. At the same time, the very slow speed of convergence and the main procedure is...
Intelligent Tutorial Systems are educational software packages that occupy Artificial Intelligence (AI) techniques and methods to represent the knowledge, as well as to conduct the learning interaction. Tutorial-like systems simulates a Socratic model of learning for teaching uncertain course material by simulating the learning process for both Teacher and a School of Students. The Student is the...
Communication bandwidth and network topology are two important factors that affect performance of distributed consensus for multi-agent systems. To deal with the limited bandwidth problem, quantization schemes have been presented. The available works about quantized average consensus require the agents to communicate in balanced digraphs, which is difficult to realize in the real world. By dropping...
Wireless sensor networks, which can achieve the target position by acquiring and processing the sensor information, has gained a widely attention in recent years. According to the localization principles with PSSI, we propose a localization method based on improved particle swarm optimization algorithm, which includes the parameters estimation of wireless signal transmission environment, the calculation...
We propose a general application programming interface called OpenATLib for auto-tuning (AT). OpenATLib is carefully designed to establish the reusability of AT functions for sparse iterative solvers. Using APIs of OpenATLib, we develop a fully auto-tuned sparse iterative solver called Xabclib. Xabclib has several novel runtime AT functions. We also develop a numerical computation policy that can...
A new family of Differential Evolution mutation strategies (DE/nrand) that are able to handle multimodal functions, have been recently proposed. The DE/nrand family incorporates information regarding the real nearest neighborhood of each potential solution, which aids them to accurately locate and maintain many global optimizers simultaneously, without the need of additional parameters. However, these...
Standard Evolution Strategy (ES) produces the next generation via the Gaussian mutation that is not directed toward the optimum. Additionally, self-adaptation mechanism is used in the standard ES to adapt mutation step-size. This paper presents a new evolution strategy which is called Quantum-inspired Evolution Strategy (QES). QES applies a new learning mechanism whereby the information of the mutants...
The mean field (MF) methods are an energy optimization method for Markov random fields (MRFs). These methods, which have their root in solid state physics, estimate the marginal density of each site of an MRF graph by iterative computation, similarly to loopy belief propagation (LBP). It appears that, being shadowed by LBP, the MF methods have not been seriously considered in the computer vision community...
An exact method for solving the problem of minimizing the maximum of a finite number of functions consists of solving a sequence of sub problems when quadratic approximations to the functions are employed in the determination of a search direction. For problems of large size, solving the sub problems exactly can be very expensive. In this paper we study truncated methods for solving the minimax problem...
We present an algorithm and implementation for distributed parallel training of single-machine multiclass SVMs. While there is ongoing and healthy debate about the best strategy for multiclass classification, there are some features of the single-machine approach that are not available when training alternatives such as one-vs-all, and that are quite complex for tree based methods. One obstacle 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.