The problem of training feedforward neural networks is considered. To solve it, new algorithms are proposed. They are based on the asymptotic analysis of the extended Kalman filter (EKF) and on a separable network str...
详细信息
The problem of training feedforward neural networks is considered. To solve it, new algorithms are proposed. They are based on the asymptotic analysis of the extended Kalman filter (EKF) and on a separable network structure. Linear weights are interpreted as diffusion random variables with zero expectation and a covariance matrix proportional to an arbitrarily large parameter lambda. Asymptotic expressions for the EKF are derived as lambda -> infinity. They are called diffusion learning algorithms (DLAs). It is shown that they are robust with respect to the accumulation of rounding errors in contrast to their prototype EKF with a large but finite lambda and that, under certain simplifying assumptions, an extreme learning machine (ELM) algorithm can be obtained from a DLA. A numerical example shows that the accuracy of a DLA may be higher than that of an ELM algorithm.
This paper belongs to neural technology and pattern recognition. It contains the description learning algorithm of back propagation. The most we pay our attention on programming algorithm.
ISBN:
(纸本)9781424445165
This paper belongs to neural technology and pattern recognition. It contains the description learning algorithm of back propagation. The most we pay our attention on programming algorithm.
The eccentricity in the Optical Disk Drive (ODD) is the inevitable deviation of the geometric center of circular tracks from the rotating center of the disk. The resulted "runout" in the drive is thus period...
详细信息
ISBN:
(纸本)0819455989
The eccentricity in the Optical Disk Drive (ODD) is the inevitable deviation of the geometric center of circular tracks from the rotating center of the disk. The resulted "runout" in the drive is thus periodic with disk rotation. To overcome the runout, conventional approach is for the pick-up head to go forward to the target track while shaking with the period runout during track accessing. This paper proposes an integration of the learning algorithm to learn the runout motion with an on-line observer to estimate the track runout during track accessing. The purpose is to allow for online computation of the target track kinematics so that the controller can adjust the accessing strategy to accommodate for the target track behavior. The proposed algorithm is demonstrated to be feasible through experiments applied to the fine jump control for a general optical storage opto-mechenical-electrical-control plant from OES in ITRI.
The inference of regular expressions from a finite number of samples has important applications in various fields such as information extraction, XML schema learning, biological sequence analysis. In this paper, we pr...
详细信息
ISBN:
(纸本)9781450366069
The inference of regular expressions from a finite number of samples has important applications in various fields such as information extraction, XML schema learning, biological sequence analysis. In this paper, we present an algorithm for learning regular expressions based on repeated string detection. The algorithm can learn a subclass of regular expressions in which unary operators such that Kleene star and Kleene plus can apply on multiple characters. Preliminary experimental results demonstrate the effectiveness and efficiency of the proposed algorithm.
With DDR5, the DRAMs will have the ability to support On Die Termination (ODT). The address topology is expected to continue to be a fly-by topology, with each DRAM loading the address bus driven by the controller. Ea...
详细信息
ISBN:
(数字)9781728142043
ISBN:
(纸本)9781728142043
With DDR5, the DRAMs will have the ability to support On Die Termination (ODT). The address topology is expected to continue to be a fly-by topology, with each DRAM loading the address bus driven by the controller. Each DRAM is expected to allow multiple ODT settings. The number of potential settings grows exponentially with the number of DRAMs in the system. Finding the optimal setting within the valid option space becomes a challenge with large number of DRAMs. At the same time, finding this optimal setting also becomes critical at the data rates of DDR5.
Fuzzy cognitive maps (FCMs) are a model for causal modeling and causal inference. It represents the real-world concepts and the causal relations between the concepts by using fuzzy variables. The major benefit of the ...
详细信息
ISBN:
(纸本)9781467376822
Fuzzy cognitive maps (FCMs) are a model for causal modeling and causal inference. It represents the real-world concepts and the causal relations between the concepts by using fuzzy variables. The major benefit of the fuzzy variables is that the model is more robust to the errors in the observed data. Although FCMs have been widely used in different research areas, it is still an open problem to efficiently construct large scale FCM models. To further improve the efficiency of the existing FCM learning algorithms, we propose a new algorithm that combines ant colony optimization algorithm, gradient descent local search and a decomposed parallel computing framework to build large scale FCMs from observational data. A set of network inference problem is used to evaluate the performance of the proposed algorithm and the results are compared to other algorithms including traditional ant colony optimization, and real coded genetic algorithms. Experimental results suggest that our algorithm outperforms the other algorithms in terms of model accuracy. We also compared the computation time required by the non-parallel ant colony optimization algorithm and the proposed parallel algorithm. When the number of nodes is appropriate, the speedup could be very close to linear speedup.
Restricted Boltzmann Machines (RBMs) is one of machine learning's methods which within past decades, the development of RBMs has quite increase. Researches of RBMs focused on theories and applications of RBMs. The...
详细信息
ISBN:
(纸本)9781450366427
Restricted Boltzmann Machines (RBMs) is one of machine learning's methods which within past decades, the development of RBMs has quite increase. Researches of RBMs focused on theories and applications of RBMs. The application of RBMs has proofed that RBMs good at finishing many tasks, such as feature extraction method, document modeling, representation learning, classification and others. The RBMs' theories also have great movements, such as the development of the learning algorithm and inference techniques of RBMs. The key factors making the RBM success on finishing task are the learning algorithm and inference techniques. They motivated the development of inference techniques which successfully improved the deep neural network (DNN) performance. The aim of this research is reviewing the various types of RBMs as the application side, and the development of learning algorithm and inference techniques as theoretical side. Hopefully, it could motivate more development on the RBMs in order to contribute on overcoming implementation tasks especially on image processing tasks.
In this paper, based on the idea of gradient descent, the learning algorithms are derived for the multi-valued neurons (MVNs) which assure the minimum of the estimation errors. In addition, finite convergence of the p...
详细信息
ISBN:
(纸本)9781728155586
In this paper, based on the idea of gradient descent, the learning algorithms are derived for the multi-valued neurons (MVNs) which assure the minimum of the estimation errors. In addition, finite convergence of the proposed learning algorithms is discussed. As an extension of the MVNs, it is finally proved that such kind of learning algorithms can also be effectively applied to universal binary neurons and multi -valued neurons with periodic activation factions, and the relating algorithms converge in finite steps.
This paper investigates the performance of Artificial Neural Network (ANN) learning algorithms for vibration-based damage detection. The capabilities of six different learning algorithms in detecting damage are studie...
详细信息
ISBN:
(纸本)9780878492060
This paper investigates the performance of Artificial Neural Network (ANN) learning algorithms for vibration-based damage detection. The capabilities of six different learning algorithms in detecting damage are studied and their performances are compared. The algorithms are Levenberg-Marquardt (LM), Resilient Backpropagation (RP), Scaled Conjugate Gradient (SCG), Conjugate Gradient with Powell-Beale Restarts (CGB), Polak-Ribiere Conjugate Gradient (CGP) and Fletcher-Reeves Conjugate Gradient (CGF) algorithms. The performances of these algorithms are assessed based on their generalisation capability in relating the vibration parameters (frequencies and mode shapes) with damage locations and severities under various numbers of input and output variables. The results show that Levenberg-Marquardt algorithm provides the best generalisation performance.
To accurately assess the driving risks associated with mixed traffic scenarios in urban areas and align with the direction of Internet of Things (IoT) technologies. An intelligent connectivity traffic safety assessmen...
详细信息
To accurately assess the driving risks associated with mixed traffic scenarios in urban areas and align with the direction of Internet of Things (IoT) technologies. An intelligent connectivity traffic safety assessment model for mixed urban traffic based on the ICVs is proposed. First, this paper proposes the traffic safety assessment model based on the driving safety field, the model integrates potential, kinetic, and behavior fields. In the process of establishing the mode, we have incorporated the acceleration parameter to dynamically capture driving risk trends. Subsequently, we define a mixed traffic scenario and calculate the driving risks for road users under different driving states based on this algorithm. The results demonstrate that the model effectively captures the driving risks of road users in different states, and the evaluation outcomes align with real-world situations, thereby validating its effectiveness. The significance of this research lies in providing a theoretical foundation for the application of the Internet of Things (IoT) in complex traffic scenarios and supporting future route planning and driving safety decision-making in intelligent transportation systems. Additionally, this model presents new ideas and methods for the development and application of ICVs technology, contributing to the advancement of intelligent transportation systems.
暂无评论