Redundant Model Research Articles

Estimation of the size at sexual maturity of the bat ray (Myliobatis californica) in northwestern Mexico through a multi-model inference

The median disc width at maturity DW^50) of males (n=91, 32.2–92.0 cm) and females (n=157, 31.0–130.0 cm) of the bat ray (Myliobatis californica) was estimated through a multi-model inference in northwestern Mexico. Gompertz’s model and four common logistics models (Lysack, Bakhayokho, White, and Brouwer and Griffiths) were compared and all fit the data well (Akaike’s differences ≤2). Logistics models estimated the same DW^50 for M. californica, and had similar Akaike’s weight suggesting that they are redundant models. Therefore, multi-model inference was performed individually with Gompertz’s model and each one of the logistic models to estimate an average model DW¯50, resulting in 64.6 cm DW for males and 99.0 cm DW for females in all analyzes. Multi-model inference is a useful tool to estimate the DW^50 of a species with greater reliability, but redundant models must not be combined in this analysis. Thus, in this case it is advisable to perform the multi-model inference with Gompertz’s model and any of logistic models. The information obtained could contribute to the fishing management of the species, which becomes more relevant considering the high percentage of immature individuals of M. californica (53% of males and 90% of females) observed in the landings from Bahía Tortugas zone, Mexico.

Active learning relevant vector machine for reliability analysis

In this study, an adaptive relevant vector machine, which is developed within a probabilistic Bayesian learning framework, is combined with Monte Carlo simulation (MCS) to perform reliability analysis with high efficiency and accuracy. A new active learning function is proposed that comprehensively considers the probability of mis-predictions and spatial locations of the existing sampling points to search for the best next point to enrich the training samples. To avoid redundant original model evaluations, an efficient stopping criterion is adopted to terminate the active learning process in a timely manner. The proposed method is tested with five examples involving non-differentiability, high non-linearity, and practical engineering problems. The proposed method can obtain highly accurate failure probability predictions with a small number of evaluations of the original model in comparison with the direct MCS and other methods in the literature. In conclusion, the proposed method is recommended for structural reliability analyses in engineering practice.

Taxonomic affiliation influences the selection of medicinal plants among people from semi-arid and humid regions-a proposition for the evaluation of utilitarian equivalence in Northeast Brazil.

BackgroundThis study sought to investigate the occurrence of taxonomic patterns between semi-arid and humid regions, verifying how the taxonomic affiliation can influence the selection of plants for medicinal purposes and act as a selection criterion.MethodsThe relationship between the taxonomic affiliation and the selection of medicinal plants with four different communities was analyzed; two of them associated with a seasonally dry tropical forest and the other two associated with a tropical rain forest. We used the Utilitarian Equivalence Model (transposing the concept of ecological equivalence, proposed by Odum, for ethnobotany) to test the hypothesis that species that have the same taxonomic affiliation tend to have the same therapeutic applications in different environments (utilitarian equivalence). In addition, we used the Utilitarian Redundancy Model to verify whether, within the same medical system, plants of the same taxonomic affiliation tend to be redundant (treating the same diseases).ResultsWe found that a pair of plants of the same genus were 9.25 times more likely to be equivalent than a different genus pair (OR = 9.25, CI [1.68–51.02], p < 0.05). When we analyzed the species used by the same population, the chances of a pair having similar therapeutic uses (utilitarian redundancy) increased when they were species of the same family (OR = 1.94, CI [1.06−3.53]; p < 0.05).ConclusionsThese findings confirm the hypothesis that there is an influence of taxonomic affiliation, in terms of genera and family, on the selection of medicinal plants in semi-arid and humid areas in Northeast Brazil. In addition, our Utilitarian Equivalence Model can be an important tool in the search for more common selection criteria, in order to identify the shared characteristics among the equivalent pairs and consequently the main types of perceptions or stimuli that led to the inclusion of such species in local pharmacopoeias.

On reliability analysis in priority standby redundant systems based on maximum entropy principle

ABSTRACT There are many studies on standby redundant systems, since those are the most standard models for ensuring reliability in systems. Particularly, two-component priority standby redundant systems are one of the most fundamental and important redundant models in the reliability theory. In recent years, the procedure for evaluating the mean time to failure (MTTF) of the two-component priority standby redundant system has been proposed by using mixed Erlang distributions as approximate distributions. In this study, by expanding the idea on evaluating MTTF proposed in the previous research, an evaluation method for the variance of failure times (VOFT) in the two-component priority standby redundant system has been presented. Furthermore, we have proposed a procedure for deriving the reliability function in the system based on the maximum entropy principle by utilizing MTTF and VOFT. Through some numerical applications, the validity of the procedures for deriving the reliability function developed in this study has been shown.

Optimal sensor placement using data-driven sparse learning method with application to pattern classification of hypersonic inlet

Placing the right amount of sensors in key locations is critical for system monitoring. In real applications, the determination of sensor placement is a compromise between monitoring performance and the costs of installation and maintenance. Given the well-interpretability of sparse learning, this paper proposes an efficient data-driven method to obtain the optimal sensor subset from the entire candidate sensor set. In order to make our model more robust to outliers and overcome the limitation of inconsistent coefficients for multiple class optimization problem, our proposed method introduces a special norm to realize the similar sparse structures of coefficients. Considering that the redundant data cannot effectively improve the real-time condition monitoring performance of engineering systems, our proposed method also includes a redundant information elimination model, which is rarely investigated in data-driven methods for optimal sensor placement problem, and this elimination model is designed by exploring the diversity of measurement data of different sensors. What’s more, we provide an alternating iteration algorithm to solve the non-smoothness convex problem of our proposed data-driven method, and the proof of its convergence has also been presented. The optimal sensor subset can be determined by the rank of the coefficients obtained by the alternating iteration algorithm. Finally, the effectiveness and feasibility of our proposed method are verified by a large number of experiments, including validation experiments on benchmark data sets and a real engineering example on the inlet model of hypersonic aircraft engine.

PARAMETER REDUNDANCY AND THE EXISTENCE OF MAXIMUM LIKELIHOOD ESTIMATES IN LOG-LINEAR MODELS

Log-linear models are typically fitted to contingency table data to describe and identify the relationship between different categorical variables. However, the data may include observed zero cell entries. The presence of zero cell entries can have an adverse effect on the estimability of parameters, due to parameter redundancy. We describe a general approach for determining whether a given log-linear model is parameter redundant for a pattern of observed zeros in the table, prior to fitting the model to the data. We derive the estimable parameters or functions of parameters and also explain how to reduce the unidentifiable model to an identifiable one. Parameter redundant models have a flat ridge in their likelihood function. We further explain when this ridge imposes some additional parameter constraints on the model, which can lead to obtaining unique maximum likelihood estimates for parameters that otherwise would not have been estimable. In contrast to other frameworks, the proposed novel approach informs on those constraints, elucidating the model that is actually being fitted.

Information Retrieval Method for Mixed Intrusion of Surveillance Network Based on Big Data

The current surveillance network intrusion information retrieval methods have problems such as low retrieval accuracy, high recall rate, and long retrieval time. Therefore, this paper proposes a hybrid intrusion information retrieval method for monitoring networks based on big data. The genetic algorithm is used to optimize the feature set, and the partial F test is introduced. The optimal subset is selected to form the optimal feature set and the redundant information elimination model is constructed to eliminate redundant information in the mixed intrusion information. Based on the information retrieval theory, the LDA model is used to build the feature set, and model the subject of the document, establish an intrusion information retrieval model, then complete the hybrid network intrusion information retrieval of the monitoring network under big data. The results show that the method proposed in this paper has higher retrieval accuracy, can effectively improve the retrieval efficiency of intrusion information and the average recall rate is about 24%, which is better than other methods.

Strategy for sensor number determination and placement optimization with incomplete information based on interval possibility model and clustering avoidance distribution index

A strategy for sensor number determination and placement optimization for structural health monitoring is proposed based on an interval possibility model and clustering avoidance distribution index. The performance of sensor configuration with optimum sensor number and less redundant information is primarily improved by considering two types of incomplete information simultaneously, namely uncertainty and redundancy. This study regards uncertainty as a non-probabilistic interval number to avoid the deficiency caused by the insufficient statistical distribution of uncertain structural parameters. As long as the bound of uncertainty is obtained, then a strategy for the sensor number can be determined using the interval possibility model and effective independence method. A novel redundant information elimination model is defined to reduce the redundant information caused by sensor clustering configuration, which considers the near neighbor distance, sensor configuration region and sensor number synchronously. Based on these two methods to address incomplete information in the sensor configuration field, the strategy for sensor placement optimization constitutes a novel combined fitness function, which comprises the effective independence method and clustering avoidance distribution index. Using the normalization and weighting factor, the combined fitness function is made more competitive by eliminating the order difference between the two objectives. Therefore, the best sensor number is suggested and the corresponding optimum sensor locations are achieved simultaneously using the genetic algorithm and the proposed strategy for sensor number and placement. The performance of sensor configuration and redundant information with incomplete information is balanced by the proposed strategy. Finally, the effectiveness of the proposed strategy and method is verified through three engineering numerical examples.

Intelligent robust control of redundant smart robotic arm Pt I: Soft computing KB optimizer - deep machine learning IT

Redundant robotic arm models as a control object discussed. Background of computational intelligence IT based on soft computing optimizer of knowledge base in smart robotic manipulators introduced. Soft computing optimizer is the toolkit of deep machine learning SW platform with optimal fuzzy neural network structure. The methods for development and design technology of intelligent control systems based on the soft computing optimizer presented in this Part 1 allow one to implement the principle of design an optimal intelligent control systems with a maximum reliability and controllability level of a complex control object under conditions of uncertainty in the source data, and in the presence of stochastic noises of various physical and statistical characters. The knowledge bases formed with the application of a soft computing optimizer produce robust control laws for the schedule of time dependent coefficient gains of conventional PID controllers for a wide range of external perturbations and are maximally insensitive to random variations of the structure of control object. The robustness of control laws is achieved by application a vector fitness function for genetic algorithm, whose one component describes the physical principle of minimum production of generalized entropy both in the control object and the control system, and the other components describe conventional control objective functionals such as minimum control error, etc. The application of soft computing technologies (Part I) for the development a robust intelligent control system that solving the problem of precision positioning redundant (3DOF and 7 DOF) manipulators considered. Application of quantum soft computing in robust intelligent control of smart manipulators in Part II described.

Modeling and simulation of FPGA-based redundant digital systems

AbstractReliability is one of the most important criteria that characterize last generation digital systems. In a wide range of applications the required reliability level is achieved by using hardware redundant configurations. Perhaps their most common form is the triple modular redundancy (TMR) based on a majority voting structure. Researchers that use this strategy make a major assumption: in fault-free operation mode the outputs of these digital systems match in all. This paper proves that synchronization and matching in all the outputs of such systems is not such a trivial problem. In this endeavor FPGA-based (Field Programmable Gate Arrays) redundant topologies are considered for study and experiments. Upon these structures specially conceived redundant models have been developed and simulated. The results outline that synchronization of complex digital systems is a difficult engineering undertaking and any initial assumption should be managed with the adequate circumspection.

A comparison of neutral genetic differentiation and genetic diversity among migratory and resident populations of Golden-crowned-Kinglets (Regulus satrapa)

Many animals migrate seasonally between breeding and non-breeding territories and these annual movements can have a profound effect on population genetic structure. We genotyped 281 individuals from 11 populations at seven variable microsatellite loci and compared patterns of neutral genetic differentiation and neutral genetic diversity among migratory and resident breeding populations of the Golden-crowned Kinglet (Regulus satrapa), a widespread North American songbird. We predicted that resident populations would exhibit greater genetic differentiation and lower genetic diversity than migratory populations because migratory behaviour is thought to enhance gene flow. The magnitude of genetic differentiation and genetic diversity among migratory and resident populations was comparable, and the greatest levels of differentiation were observed for pairwise comparisons between Ontario and all western populations. Distance-based redundancy models and redundancy models revealed that patterns of neutral genetic differentiation and neutral genetic diversity follow an isolation-by-distance model and are not correlated with migratory behaviour. Overall it appears that genetic patterns are more closely associated with Pleistocene glacial history as proposed in a previous study.

Representation-supporting model elements

It is assumed that scientific models contain no superfluous model elements in scientific representation. A representational model is constructed with all the model elements serving the representational purpose. The received view has it that there are no redundant model elements which are non-representational. Contrary to this received view, I argue that there exist some non-representational model elements which are essential in scientific representation. I call them representation-supporting model elements in virtue of the fact that they play the role to support the representational role of representational model elements. Representation-supporting model elements, which have a characteristic of dependability, are not to be eliminated in the process of modeling although they are playing second fiddle to representational model elements in scientific representation.

Accurate Uncertainty Propagation for Discrete-Time Nonlinear Systems Using Differential Inequalities With Model Redundancy

This article presents new methods for computing tight interval enclosures of the reachable sets of discrete-time nonlinear systems subject to bounded uncertainties. These methods are motivated by effective methods for continuous-time systems based on differential inequalities (DI). Modern DI methods, particularly those using refinement techniques based on redundant model equations, have demonstrated very sharp enclosures at low cost for several challenging test cases. However, they rely on key properties of continuous-time systems that do not hold generally in discrete time. Nevertheless, we show that discrete-time analogues of these methods do provide valid enclosures for systems satisfying certain monotonicity conditions. We show that these conditions are always satisfied for systems obtained by forward Euler discretization of continuous-time models with step sizes below a computable upper limit. New refinement algorithms are presented for exploiting redundant model equations in the discrete-time setting. The resulting discrete-time DI methods are compared to existing algorithms using several case studies.

Structured Pruning for Efficient Convolutional Neural Networks via Incremental Regularization

Modern Convolutional Neural Networks (CNNs) are usually restricted by their massive computation and high storage. Parameter pruning is a promising approach for CNN compression and acceleration by eliminating redundant model parameters with tolerable performance degradation. Despite its effectiveness, existing regularization-based parameter pruning methods usually drive weights towards zero with large and constant regularization factors, which neglects the fragility of the expressiveness of CNNs, and thus calls for a more gentle regularization scheme so that the networks can adapt during pruning. To achieve this, we propose a novel regularization-based pruning method, named IncReg , to incrementally assign different regularization factors to different weights based on their relative importance. Empirical analysis on CIFAR-10 dataset verifies the merits of IncReg. Further extensive experiments with popular CNNs on CIFAR-10 and ImageNet datasets show that IncReg achieves comparable to even better results compared with state-of-the-arts. Moreover, to resolve the problem that column pruning cannot be directly applied to off-the-shelf deep learning libraries for acceleration, we generalize IncReg from column pruning to spatial pruning, which can equip existing structured pruning methods (such as channel pruning) for further acceleration with ignorable accuracy loss. Our source codes and trained models are available at: https://github.com/mingsun-tse/caffe_increghttps://github.com/mingsun-tse/caffe_increg .

Two redundant rule based algorithms for time-delay nonlinear models: least squares iterative and particle swarm optimisation

Two redundant rule based methods are developed for a time-delay nonlinear model in this paper. By using the redundant rule, the time-delay nonlinear model can be turned into a redundant model which contains some redundant terms. Then the least squares iterative and the particle swarm optimisation algorithms are applied to update the parameters and the corresponding time-delay. Compared with the redundant rule based least squares iterative algorithm, the redundant rule based particle swarm optimisation algorithm is more efficient for nonlinear models with complex structures. A simulation example shows that the proposed algorithms are effective.

Redundant Iaas Cloud Selection With Consideration Of Multi Criteria Decision Analysis

Due to the rapid growth of dependency over Mobile cloud system, the demand of high computational system increases rapidly. Cloud computing gives the flexibility to users to use high computation system at nominal cost and scalability in dynamic and on demand fashion. The term offloading has attracted the researcher to obtain the highly capable cloud system with certain limitations such as minimum processing and communication time, make span, minimum operational cost. Offloading of data and application have definite positive keynotes such as it can extend the battery life of IOT devices also it is suitable for critical events (events those require minimum response time). In today, numerous cloud services providers are offering customized services, they are dedicated to fulfill the demands of user with negotiable service level agreement. But due to the inherent uncertainty involved in human judgment and lack of learning capacity, a dynamic cloud selection and decision model is required to evaluate the user preferences. That can recommend an optimal and redundant cloud system from the available pool of cloud service providers. Resolving of uncertainties and ambiguity in human’s decision are solved through fuzzy set theory. In this paper, an optimal and redundant cloud selection model has been presented on the basis of multi criteria decision analysis under consideration. Weighted Sum Model, Fuzzy Analytic Hierarchy Process and Fuzzy Revised Analytic Hierarchy Process are evaluated on 10 different criterions. Overall the outranking result for the considered datasets is similar, while the computation power of AHP method is ideally superior with comparison to revised AHP method.

Neutrosophic Logic for Software Effort Estimation: An Improvement over Fuzzy Logic

Software effort estimation is big and active research area. Software effort estimation is useful for time and efforts required to perform a particular task. But, it is very rare to estimate the effort with high level of reliability. There are various approaches to estimate the software application effort. In the present paper, to estimate the effort for software applications efforts, neutrosophic logic approach is used. Neutrosophic logic is a mathematical model for ambiguity, uncertainty, incompleteness, vagueness, redundancy, contradiction and inconsistency in data. It is the extension to the fuzzy logic. It is capable of handling those errors which are not handled by fuzzy logic like indeterminacy in the data. Neutrosophic logic gives the results very similar to human thinking. The present work concludes that neutrosophic logic optimizes the performance of fuzzy logic while calculating the software efforts..

A vanishing moment ensemble model for wind speed multi-step prediction with multi-objective base model selection

Wind speed forecasting model can improve the safety and economy of wind energy utilization. In this study, a new hybrid ensemble model is presented. In the proposed ensemble model, the base models are the wavelet-based models with 10 different vanishing moments. The base models are selected by multi-objective feasibility enhanced particle swarm optimization algorithm to maximize relevancy and minimize redundancy. In this manner, the redundant base models can be discarded, while enhancing the relevancy between the forecasting results of the base models and actual data. Then, the ensemble weights of the selected base models are calculated by non-dominated sorting genetic algorithm III. The optimization objective function is optimizing the accuracy and robustness of the ensemble model. Three real wind speed series collected from three different sites in Xinjiang are utilized for data simulation. The experimental results show that: (1) the proposed hybrid model can generate high-accuracy wind speed forecasting results in the studied cases; (2) the proposed hybrid model can outperform other benchmark models in the studied cases.

Redundancy with Processor Sharing servers

The main motivation to investigate redundancy models comes from empirical evidence suggesting that redundancy can help improve the performance of real-world applications. Under redundancy, a job that arrives to this system is dispatched to d servers uniformly chosen at random in order to benefit from the variability of the length of these queues. As soon as one of the copies finishes service, the job (and its copies) is removed from the system, and as a consequence, a job's delay is given by the minimum delay among the servers its copies are sent to. Most of the literature on performance evaluation of redundancy systems has been carried out when First Come First Served (FCFS) is implemented in the servers. In particular, for exponential service time distributions, Gardner et al. [4, 5] and Bonald and Comte [2] show that the stability region is not reduced due to adding redundant copies. In this extended abstract, we focus instead on Processor Sharing (PS) service policy and study how redundancy impacts the stability condition. In particular, we aim to study the impact that the correlation structure of the copies has on the performance of the redundancy-d model. In a recent paper, Gardner et al. [3] showed that the assumption of independent and identically distributed (i.i.d.) copies, can be unrealistic, and that it might lead to theoretical results that do not reflect the results of replication schemes in real-life computer systems. We consider the two extreme cases of correlation; (i) the copies are i.i.d. (ii) the copies of a job are exact replicas (identical copies). We observe that the stability condition strongly depends on the correlation structure, as well as on the number of redundant copies.

HbtSOTLearner: A Hierarchical-Backtracking-Based Parameter Learner for Small Outline Transistor

This paper is concerned with parameter learning of chips with small outline transistor (SOT) package, which is one of the most widely used package in surface mount technology (SMT) and has various subcategories. Previously learned parameter is crucial to most SOT-related industrial applications, such as location and defect inspection. However, parameter learning is a challenging work because of package diversity and image-quality deterioration in practical industrial applications. The conventional methods, checking data sheet or manual measuring, cannot meet the accuracy requirement of SMT. This paper proposes a hierarchical-backtracking-based parameter learner for SOT chips. The Gaussian mixture model based clustering algorithm and random walker algorithm are firstly applied to extracting lead regions of SOT chip; Then, chip models are inferred by grouping these lead regions with a hierarchical backtracking algorithm. Finally, redundant models are eliminated with root set pyramids and the valid chip model is obtained. The experimental results show that the proposed parameter learner performs well on SOT chips and is robust to noisy sets.