Journal articles on the topic 'Weight adaptation algorithms'
To see the other types of publications on this topic, follow the link: Weight adaptation algorithms.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Weight adaptation algorithms.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Zhong, Shuiming, Yu Xue, Yunhao Jiang, Yuanfeng Jin, Jing Yang, Ping Yang, Yuan Tian, and Mznah Al-Rodhaan. "A Sensitivity-Based Improving Learning Algorithm for Madaline Rule II." Mathematical Problems in Engineering 2014 (2014): 1–8. http://dx.doi.org/10.1155/2014/219679.
Full textAbstract:
This paper proposes a new adaptive learning algorithm for Madalines based on a sensitivity measure that is established to investigate the effect of a Madaline weight adaptation on its output. The algorithm, following the basic idea of minimal disturbance as the MRII did, introduces an adaptation selection rule by means of the sensitivity measure to more accurately locate the weights in real need of adaptation. Experimental results on some benchmark data demonstrate that the proposed algorithm has much better learning performance than the MRII and the BP algorithms.
2
Magoulas, G. D., M. N. Vrahatis, and G. S. Androulakis. "Improving the Convergence of the Backpropagation Algorithm Using Learning Rate Adaptation Methods." Neural Computation 11, no. 7 (October 1, 1999): 1769–96. http://dx.doi.org/10.1162/089976699300016223.
Full textAbstract:
This article focuses on gradient-based backpropagation algorithms that use either a common adaptive learning rate for all weights or an individual adaptive learning rate for each weight and apply the Goldstein/Armijo line search. The learning-rate adaptation is based on descent techniques and estimates of the local Lipschitz constant that are obtained without additional error function and gradient evaluations. The proposed algorithms improve the backpropagation training in terms of both convergence rate and convergence characteristics, such as stable learning and robustness to oscillations. Simulations are conducted to compare and evaluate the convergence behavior of these gradient-based training algorithms with several popular training methods.
3
Yang, Feiran, Yin Cao, Ming Wu, Felix Albu, and Jun Yang. "Frequency-Domain Filtered-x LMS Algorithms for Active Noise Control: A Review and New Insights." Applied Sciences 8, no. 11 (November 20, 2018): 2313. http://dx.doi.org/10.3390/app8112313.
Full textAbstract:
This paper presents a comprehensive overview of the frequency-domain filtered-x least mean-square (FxLMS) algorithms for active noise control (ANC). The direct use of frequency-domain adaptive filters for ANC results in two kinds of delays, i.e., delay in the signal path and delay in the weight adaptation. The effects of the two kinds of delays on the convergence behavior and stability of the adaptive algorithms are analyzed in this paper. The first delay can violate the so-called causality constraint, which is a major concern for broadband ANC, and the second delay can reduce the upper bound of the step size. The modified filter-x scheme has been employed to remove the delay in the weight adaptation, and several delayless filtering approaches have been presented to remove the delay in the signal path. However, state-of-the-art frequency-domain FxLMS algorithms only remove one kind of delay, and some of these algorithms have a very high peak complexity and hence are impractical for real-time systems. This paper thus proposes a new delayless frequency-domain ANC algorithm that completely removes the two kinds of delays and has a low complexity. The performance advantages and limitations of each algorithm are discussed based on an extensive evaluation, and the complexities are evaluated in terms of both the peak and average complexities.
4
Cufoglu, Ayse, Mahi Lohi, and Colin Everiss. "Feature weighted clustering for user profiling." International Journal of Modeling, Simulation, and Scientific Computing 08, no. 04 (December 2017): 1750056. http://dx.doi.org/10.1142/s1793962317500568.
Full textAbstract:
Personalization is the adaptation of the services to fit the user’s interests, characteristics and needs. The key to effective personalization is user profiling. Apart from traditional collaborative and content-based approaches, a number of classification and clustering algorithms have been used to classify user related information to create user profiles. However, they are not able to achieve accurate user profiles. In this paper, we present a new clustering algorithm, namely Multi-Dimensional Clustering (MDC), to determine user profiling. The MDC is a version of the Instance-Based Learner (IBL) algorithm that assigns weights to feature values and considers these weights for the clustering. Three feature weight methods are proposed for the MDC and, all three, have been tested and evaluated. Simulations were conducted with using two sets of user profile datasets, which are the training (includes 10,000 instances) and test (includes 1000 instances) datasets. These datasets reflect each user’s personal information, preferences and interests. Additional simulations and comparisons with existing weighted and non-weighted instance-based algorithms were carried out in order to demonstrate the performance of proposed algorithm. Experimental results using the user profile datasets demonstrate that the proposed algorithm has better clustering accuracy performance compared to other algorithms. This work is based on the doctoral thesis of the corresponding author.
5
Wu, Di, Sheng Yao Yang, and J. C. Liu. "Cognitive Radio Decision Engine Based on Multi-Objective Genetic Algorithm." Applied Mechanics and Materials 48-49 (February 2011): 314–17. http://dx.doi.org/10.4028/www.scientific.net/amm.48-49.314.
Full textAbstract:
The performance optimization of cognitive radio is a multi-objective optimization problem. Existing genetic algorithms are difficult to assign the weight of each objective when the linear weighting method is used to simplify the multi-objective optimization problem into a single objective optimization problem. In this paper, we propose a new cognitive decision engine algorithm using multi-objective genetic algorithm with population adaptation. A multicarrier system is used for simulation analysis, and experimental results show that the proposed algorithm is effective and meets the real-time requirement.
6
Yiou, Pascal, and Aglaé Jézéquel. "Simulation of extreme heat waves with empirical importance sampling." Geoscientific Model Development 13, no. 2 (February 25, 2020): 763–81. http://dx.doi.org/10.5194/gmd-13-763-2020.
Full textAbstract:
Abstract. Simulating ensembles of extreme events is a necessary task to evaluate their probability distribution and analyze their meteorological properties. Algorithms of importance sampling have provided a way to simulate trajectories of dynamical systems (like climate models) that yield extreme behavior, like heat waves. Such algorithms also give access to the return periods of such events. We present an adaptation based on circulation analogues of importance sampling to provide a data-based algorithm that simulates extreme events like heat waves in a realistic way. This algorithm is a modification of a stochastic weather generator, which gives more weight to trajectories with higher temperatures. This presentation outlines the methodology using European heat waves and illustrates the spatial and temporal properties of simulations.
7
Khabarlak, K. S. "FASTER OPTIMIZATION-BASED META-LEARNING ADAPTATION PHASE." Radio Electronics, Computer Science, Control, no. 1 (April 7, 2022): 82. http://dx.doi.org/10.15588/1607-3274-2022-1-10.
Full textAbstract:
Context. Neural networks require a large amount of annotated data to learn. Meta-learning algorithms propose a way to decrease number of training samples to only a few. One of the most prominent optimization-based meta-learning algorithms is MAML. However, its adaptation to new tasks is quite slow. The object of study is the process of meta-learning and adaptation phase as defined by the MAML algorithm.Objective. The goal of this work is creation of an approach, which should make it possible to: 1) increase the execution speed of MAML adaptation phase; 2) improve MAML accuracy in certain cases. The testing results will be shown on a publicly available few-shot learning dataset CIFAR-FS.Method. In this work an improvement to MAML meta-learning algorithm is proposed. Meta-learning procedure is defined in terms of tasks. In case of image classification problem, each task is to try to learn to classify images of new classes given only a few training examples. MAML defines 2 stages for the learning procedure: 1) adaptation to the new task; 2) meta-weights update. The whole training procedure requires Hessian computation, which makes the method computationally expensive. After being trained, the network will typically be used for adaptation to new tasks and the subsequent prediction on them. Thus, improving adaptation time is an important problem, which we focus on in this work. We introduce lambda pattern by which we restrict which weight we update in the network during the adaptation phase. This approach allows us to skip certain gradient computations. The pattern is selected given an allowed quality degradation threshold parameter. Among the pattern that fit the criteria, the fastest pattern is then selected. However, as it is discussed later, quality improvement is also possible is certain cases by a careful pattern selection.Results. The MAML algorithm with lambda pattern adaptation has been implemented, trained and tested on the open CIFAR-FS dataset. This makes our results easily reproducible.Conclusions. The experiments conducted have shown that via lambda adaptation pattern selection, it is possible to significantly improve the MAML method in the following areas: adaptation time has been decreased by a factor of 3 with minimal accuracy loss. Interestingly, accuracy for one-step adaptation has been substantially improved by using lambda patterns as well. Prospects for further research are to investigate a way of a more robust automatic pattern selection scheme.
8
Li, Miqing, and Xin Yao. "What Weights Work for You? Adapting Weights for Any Pareto Front Shape in Decomposition-Based Evolutionary Multiobjective Optimisation." Evolutionary Computation 28, no. 2 (June 2020): 227–53. http://dx.doi.org/10.1162/evco_a_00269.
Full textAbstract:
The quality of solution sets generated by decomposition-based evolutionary multi-objective optimisation (EMO) algorithms depends heavily on the consistency between a given problem's Pareto front shape and the specified weights' distribution. A set of weights distributed uniformly in a simplex often leads to a set of well-distributed solutions on a Pareto front with a simplex-like shape, but may fail on other Pareto front shapes. It is an open problem on how to specify a set of appropriate weights without the information of the problem's Pareto front beforehand. In this article, we propose an approach to adapt weights during the evolutionary process (called AdaW). AdaW progressively seeks a suitable distribution of weights for the given problem by elaborating several key parts in weight adaptation—weight generation, weight addition, weight deletion, and weight update frequency. Experimental results have shown the effectiveness of the proposed approach. AdaW works well for Pareto fronts with very different shapes: 1) the simplex-like, 2) the inverted simplex-like, 3) the highly nonlinear, 4) the disconnect, 5) the degenerate, 6) the scaled, and 7) the high-dimensional.
9
Miertoiu, Florin Ilarion, and Bogdan Dumitrescu. "Feasibility Pump Algorithm for Sparse Representation under Gaussian Noise." Algorithms 13, no. 4 (April 9, 2020): 88. http://dx.doi.org/10.3390/a13040088.
Full textAbstract:
In this paper, the Feasibility Pump is adapted for the problem of sparse representations of signals affected by Gaussian noise. This adaptation is tested and then compared to Orthogonal Matching Pursuit (OMP) and the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA). The feasibility pump recovers the true support much better than the other two algorithms and, as the SNR decreases and the support size increases, it has a smaller recovery and representation error when compared with its competitors. It is observed that, in order for the algorithm to be efficient, a regularization parameter and a weight term for the error are needed.
10
Szuster, Marcin, and Piotr Gierlak. "Globalized Dual Heuristic Dynamic Programming in Control of Robotic Manipulator." Applied Mechanics and Materials 817 (January 2016): 150–61. http://dx.doi.org/10.4028/www.scientific.net/amm.817.150.
Full textAbstract:
The article focuses on the implementation of the globalized dual-heuristic dynamic programming algorithm in the discrete tracking control system of the three degrees of freedom robotic manipulator. The globalized dual-heuristic dynamic programming algorithm is included in the approximate dynamic programming algorithms family, that bases on the Bellman’s dynamic programming idea. These algorithms generally consist of the actor and the critic structures realized in a form of artificial neural networks. Moreover, the control system includes the PD controller, the supervisory term and an additional control signal. The structure of the supervisory term derives from the stability analysis, which was realized using the Lyapunov stability theorem. The control system works on-line and the neural networks’ weight adaptation process is realized in every iteration step. A series of computer simulations was realized in Matlab/Simulink software to confirm performance of the control system.
11
OWAIS, M. A., and M. A. AHMED. "DYNAMIC SIMILARITY METRIC USING FUZZY PREDICATES FOR CASE-BASED PLANNING." International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems 17, no. 01 (February 2009): 47–68. http://dx.doi.org/10.1142/s0218488509005735.
Full textAbstract:
Case-based planning (CBP) is a knowledge-based planning technique which develops new plans by reusing its past experience instead of planning from scratch. The task of CBP becomes difficult when the knowledge needed for planning can not be expressed precisely. In this paper, we tackle this issue by modeling imprecise information using fuzzy predicates; and accordingly, we present a dynamic similarity metric for efficient and effective retrieval of relevant cases from a library of cases. We also present weight adaptation algorithm to allow improving the performance of the metric overtime. We use and compare the performance of Tabu search, simulated annealing, and exhaustive search algorithms in instantiating fuzzy predicates to achieve maximum similarity between a new problem and a case. Our experiments show that the proposed metric is sound. The metric along with the adaptation algorithm have been shown to be promising when compared to others. Experiments also show that simulated annealing is more efficient than Tabu search and exhaustive search in fuzzy predicates instantiation.
12
Valdez, Fevrier, Juan Carlos Vazquez, and Fernando Gaxiola. "Fuzzy Dynamic Parameter Adaptation in ACO and PSO for Designing Fuzzy Controllers: The Cases of Water Level and Temperature Control." Advances in Fuzzy Systems 2018 (July 2, 2018): 1–19. http://dx.doi.org/10.1155/2018/1274969.
Full textAbstract:
A novel approach applied to Particle Swarm Optimization (PSO) and Ant Colony Optimization is presented. The main contribution of this work is the use of fuzzy systems to dynamically update the parameters for the ACO and PSO algorithms. In the case of ACO, two fuzzy systems are designed for the Ant Colony System (ACS) algorithm variant. The first system adjusts the value for the pheromone evaporation parameter from the global pheromone trail update equation and the second system adjusts the values for the pheromone evaporation parameter from the local pheromone trail update equation. In the case of PSO, a fuzzy system is designed to find the values for the inertia weight parameter from the velocity equation. Fuzzy logic controllers (FLCs) are optimized with ACO and PSO, respectively, to prove the performance of the proposed approach. The particular benchmark problems considered to test the proposed methods are the water level control in a tank and temperature control in a shower. Therefore, PSO and ACO algorithms are applied in the optimization of the parameters of the FLCs. The achievement of the proposed fuzzy ACO and PSO algorithms is compared with the original results of each benchmark control problem.
13
Piza, D. M., D. S. Semenov, and S. V. Morshchavka. "Efficiency Estimation of Discrete Algorithms for Adaptation of Weight Coefficients in Space-Time Processing of Radar Signals." Radioelectronics and Communications Systems 62, no. 1 (January 2019): 6–11. http://dx.doi.org/10.3103/s0735272719010023.
Full text
14
Yan, Xue Song, Qing Hua Wu, Cheng Yu Hu, and Qing Zhong Liang. "Research of Space Electronic Circuit Optimization Design." Applied Mechanics and Materials 48-49 (February 2011): 932–36. http://dx.doi.org/10.4028/www.scientific.net/amm.48-49.932.
Full textAbstract:
During the space electronic system in carries out the exploratory mission in the deep space, it maybe faced with kinds of violent natural environment, to electric circuit's performance, the volume, the weight and the stability proposed a higher request, the traditional circuit design method already more and more with difficulty satisfied this kind of request. The traditional circuit design method already more and more with difficulty satisfied this kind of request. But unifies the programmable component and the evolutionary algorithms hardware may the dynamic change hardware's structure adapt the adverse circumstance, resume the damage of the function, the adaptation for the duty change. In view of the Xilinx Company's FPGA unique feature, proposed one kind of evolutionary algorithms which uses in the space electronic system circuit optimization design and through the experiment proved, the algorithm obtains the circuit structure to surpass the tradition circuit design method.
15
Yan, Xue Song, Qing Hua Wu, Cheng Yu Hu, and Qing Zhong Liang. "Circuit Optimization Design Using Evolutionary Algorithms." Advanced Materials Research 187 (February 2011): 303–8. http://dx.doi.org/10.4028/www.scientific.net/amr.187.303.
Full textAbstract:
During the space electronic system in carries out the exploratory mission in the deep space, it maybe faced with kinds of violent natural environment, to electric circuit's performance, the volume, the weight and the stability proposed a higher request, the traditional circuit design method already more and more with difficulty satisfied this kind of request. The traditional circuit design method already more and more with difficulty satisfied this kind of request. But unifies the programmable component and the evolutionary algorithms hardware may the dynamic change hardware's structure adapt the adverse circumstance, resume the damage of the function, the adaptation for the duty change. After the optimization, obtains the circuit structure will often stem from our anticipation, this will be the altitude which the experience and the skillful institute hope to attain with difficulty. In view of the Xilinx Company's FPGA unique feature, proposed one kind of evolutionary algorithms which uses in the space electronic system circuit optimization design and through the experiment proved, the algorithm obtains the circuit structure to surpass the tradition circuit design method. This work investigates the application of genetic algorithms in the field of circuit optimization. For the case studies, this means has proved to be efficient and the experiment results show that the new means have got the better results.
16
Kewalramani, Manish A., and Rajiv Gupta. "Group Method of Data Handling Algorithms to Predict Compressive Strength of Concrete Based on Absorbed Extraterrestrial Solar Radiations." Key Engineering Materials 689 (April 2016): 108–13. http://dx.doi.org/10.4028/www.scientific.net/kem.689.108.
Full textAbstract:
The present study applies group method of data handling (GMDH) to predict compressive strength of normal strength concrete based on experimentally determined weight, ultrasonic pulse velocity and extraterrestrial solar radiations absorbed by concrete specimen. GMDH are widely used as mathematical modelling and non-linear regression algorithms, and are assumed as specific type of supervised artificial neural networks. Concrete being a multi-phase porous and non-linear material justifies usage of such algorithm as GMDH employs the idea of natural selection to control size, complexity and accuracy of networks being used for various applications like function approximation, non-linear regression and pattern recognition. The effectiveness of algorithm is validated when 60%, 70%, 80% and 100% of normalized and non-normalized data is used for training. GMDH being an intelligent algorithm with ability of learning and adaptation can be conveniently used as an appropriate prediction tool for non-linear complex systems like concrete.
17
Zhao, Sicheng, Guangzhi Wang, Shanghang Zhang, Yang Gu, Yaxian Li, Zhichao Song, Pengfei Xu, Runbo Hu, Hua Chai, and Kurt Keutzer. "Multi-Source Distilling Domain Adaptation." Proceedings of the AAAI Conference on Artificial Intelligence 34, no. 07 (April 3, 2020): 12975–83. http://dx.doi.org/10.1609/aaai.v34i07.6997.
Full textAbstract:
Deep neural networks suffer from performance decay when there is domain shift between the labeled source domain and unlabeled target domain, which motivates the research on domain adaptation (DA). Conventional DA methods usually assume that the labeled data is sampled from a single source distribution. However, in practice, labeled data may be collected from multiple sources, while naive application of the single-source DA algorithms may lead to suboptimal solutions. In this paper, we propose a novel multi-source distilling domain adaptation (MDDA) network, which not only considers the different distances among multiple sources and the target, but also investigates the different similarities of the source samples to the target ones. Specifically, the proposed MDDA includes four stages: (1) pre-train the source classifiers separately using the training data from each source; (2) adversarially map the target into the feature space of each source respectively by minimizing the empirical Wasserstein distance between source and target; (3) select the source training samples that are closer to the target to fine-tune the source classifiers; and (4) classify each encoded target feature by corresponding source classifier, and aggregate different predictions using respective domain weight, which corresponds to the discrepancy between each source and target. Extensive experiments are conducted on public DA benchmarks, and the results demonstrate that the proposed MDDA significantly outperforms the state-of-the-art approaches. Our source code is released at: https://github.com/daoyuan98/MDDA.
18
Meng, Yu, Jianping Xuan, Long Xu, and Jie Liu. "Dynamic Reweighted Domain Adaption for Cross-Domain Bearing Fault Diagnosis." Machines 10, no. 4 (March 30, 2022): 245. http://dx.doi.org/10.3390/machines10040245.
Full textAbstract:
Certain progress has been made in fault diagnosis under cross-domain scenarios recently. Most researchers have paid almost all their attention to promoting domain adaptation in a common space. However, several challenges that will cause negative transfer have been ignored. In this paper, a reweighting method is proposed to overcome this difficulty from two aspects. First, extracted features differ greatly from one another in promoting positive transfer, and measuring the difference is important. Measured by conditional entropy, the weight of adversarial losses for those well aligned features are reduced. Second, the balance between domain adaptation and class discrimination greatly influences the transferring task. Here, a dynamic weight strategy is adopted to compute the balance factor. Consideration is made from the perspective of maximum mean discrepancy and multiclass linear discriminant analysis. The first item is supposed to measure the degree of the domain adaptation between source and the target domain, and the second is supposed to show the classification performance of the classifier on the learned features in the current training epoch. Finally, extensive experiments on several bearing fault diagnosis datasets are conducted. The performance shows that our model has an obvious advantage compared with other common transferring algorithms.
19
Lin, Xianghong, Mengwei Zhang, and Xiangwen Wang. "Supervised Learning Algorithm for Multilayer Spiking Neural Networks with Long-Term Memory Spike Response Model." Computational Intelligence and Neuroscience 2021 (November 24, 2021): 1–16. http://dx.doi.org/10.1155/2021/8592824.
Full textAbstract:
As a new brain-inspired computational model of artificial neural networks, spiking neural networks transmit and process information via precisely timed spike trains. Constructing efficient learning methods is a significant research field in spiking neural networks. In this paper, we present a supervised learning algorithm for multilayer feedforward spiking neural networks; all neurons can fire multiple spikes in all layers. The feedforward network consists of spiking neurons governed by biologically plausible long-term memory spike response model, in which the effect of earlier spikes on the refractoriness is not neglected to incorporate adaptation effects. The gradient descent method is employed to derive synaptic weight updating rule for learning spike trains. The proposed algorithm is tested and verified on spatiotemporal pattern learning problems, including a set of spike train learning tasks and nonlinear pattern classification problems on four UCI datasets. Simulation results indicate that the proposed algorithm can improve learning accuracy in comparison with other supervised learning algorithms.
20
Gosztonyi, Márton, and Csákné Filep Judit. "Profiling (Non-)Nascent Entrepreneurs in Hungary Based on Machine Learning Approaches." Sustainability 14, no. 6 (March 18, 2022): 3571. http://dx.doi.org/10.3390/su14063571.
Full textAbstract:
In our study, we examined the characteristics of nascent entrepreneurs using the 2021 Global Entrepreneurship Monitor national representative data in Hungary. We examined our topic based on Arenius and Minitti’s four-category theory framework. In our research, we examined system-level feature sets with four machine learning modeling algorithms: multivariate adaptive regression spline (MARS), support vector machine (SVM), random forest (RF), and AdaBoost. Our results show that each machine algorithm can predict nascent entrepreneurs with over 90% adaptive cruise control (ACC) accuracy. Furthermore, the adaptation of the categories of variables based on the theory of Arenius and Minitti provides an appropriate framework for obtaining reliable predictions. Based on our results, it can be concluded that perceptual factors have different importance and weight along the optimal models, and if we include further reliability measures in the model validation, we cannot pinpoint only one algorithm that can adequately identify nascent entrepreneurs. Accurate forecasting requires a careful and predictor-level analysis of the algorithms’ models, which also includes the systemic relationship between the affecting factors. An important but unexpected result of our study is that we identified that Hungarian NEs have very specific previous entrepreneurial and business ownership experience; thus, they can be defined not as a beginner but as a novice enterprise.
21
Khunteta, Ajay, and D. Ghosh. "Object Boundary Detection Using Active Contour Model via Multiswarm PSO with Fuzzy-Rule Based Adaptation of Inertia Factor." Advances in Fuzzy Systems 2016 (2016): 1–20. http://dx.doi.org/10.1155/2016/6179576.
Full textAbstract:
Active contour models, colloquially known as snakes, are quite popular for several applications such as object boundary detection, image segmentation, object tracking, and classification via energy minimization. While energy minimization may be accomplished using traditional optimization methods, approaches based on nature-inspired evolutionary algorithms have been developed in recent years. One such evolutionary algorithm that has been used extensively in active contours is the particle swarm optimization (PSO). However, conventional PSO converges slowly and gets trapped in local minimum easily which results in inaccurate detection of concavities in the object boundary. This is taken care of by using proposed multiswarm PSO in which a swarm is set for every control point in the snake and then all the swarms search for their best points simultaneously through information sharing among them. The performance of the multiswarm PSO-based search process is further enhanced by using dynamic adaptation of the inertia factor. In this paper, we propose using a set of fuzzy rules to adjust the inertia weight on the basis of the current normalized snake energy and the current value of inertia. Experimental results demonstrate the effectiveness of the proposed method compared to conventional approaches.
22
Lightheart, Toby, Steven Grainger, and Tien-Fu Lu. "Spike-Timing-Dependent Construction." Neural Computation 25, no. 10 (October 2013): 2611–45. http://dx.doi.org/10.1162/neco_a_00501.
Full textAbstract:
Spike-timing-dependent construction (STDC) is the production of new spiking neurons and connections in a simulated neural network in response to neuron activity. Following the discovery of spike-timing-dependent plasticity (STDP), significant effort has gone into the modeling and simulation of adaptation in spiking neural networks (SNNs). Limitations in computational power imposed by network topology, however, constrain learning capabilities through connection weight modification alone. Constructive algorithms produce new neurons and connections, allowing automatic structural responses for applications of unknown complexity and nonstationary solutions. A conceptual analogy is developed and extended to theoretical conditions for modeling synaptic plasticity as network construction. Generalizing past constructive algorithms, we propose a framework for the design of novel constructive SNNs and demonstrate its application in the development of simulations for the validation of developed theory. Potential directions of future research and applications of STDC for biological modeling and machine learning are also discussed.
23
Rincón-García, Eric Alfredo, Miguel Ángel Gutiérrez-Andrade, Sergio Gerardo de-los-Cobos-Silva, Roman Anselmo Mora-Gutiérrez, Antonin Ponsich, and Pedro Lara-Velázquez. "A comparative study of population-based algorithms for a political districting problem." Kybernetes 46, no. 1 (January 9, 2017): 172–90. http://dx.doi.org/10.1108/k-06-2016-0130.
Full textAbstract:
Purpose This paper aims to propose comparing the performance of three algorithms based on different population-based heuristics, particle swarm optimization (PSO), artificial bee colony (ABC) and method of musical composition (DMMC), for the districting problem. Design/methodology/approach In order to compare the performance of the proposed algorithms, they were tested on eight instances drawn from the Mexican electoral institute database, and their respective performance levels were compared. In addition, a simulated annealing-based (simulated annealing – SA) algorithm was used as reference to evaluate the proposed algorithms. This technique was included in this work because it has been used for Federal districting in Mexico since 1994. The performance of the algorithms was evaluated in terms of the quality of the approximated Pareto front and efficiency. Regarding solution quality, convergence and dispersion of the resulting non-dominated solutions were evaluated. Findings The results show that the quality and diversification of non-dominated solutions generated by population-based algorithms are better than those produced by Federal Electoral Institute’s (IFE’s) SA-based technique. More accurately, among population-based techniques, discrete adaptation of ABC and MMC outperform PSO. Originality/value The performance of three population-based techniques was evaluated for the districting problem. In this paper, the authors used the objective function proposed by the Mexican IFE, a weight aggregation function that seeks for a districting plan that represents the best balance between population equality and compactness. However, the weighting factors can be modified by political agreements; thus, the authors decided to produce a set of efficient solutions, using different weighting factors for the computational experiments. This way, the best algorithm will produce high quality solutions no matter the weighting factors used for a real districting process. The computational experiments proved that the proposed artificial bee colony and method of musical composition-based algorithms produce better quality efficient solutions than its counterparts. These results show that population-based algorithms can outperform traditional local search strategies. Besides, as far as we know, this is the first time that the method of musical composition is used for this kind of problems.
24
Nitsuk, Yu A., О. М. Semchak, and І. V. Sharipova. "WAYS OF DIMINISHING OF ERRORS OF CALCULATIONS OF COMPUTER OF AUTONOMOUS MOBILE OBJECT ARE FOR ALGORITHMS OF SLAM OF NAVIGATION." Проблеми створення, випробування, застосування та експлуатації складних інформаційних систем, no. 18 (December 30, 2020): 32–43. http://dx.doi.org/10.46972/2076-1546.2020.18.04.
Full textAbstract:
A question is in-process considered, in relation to the lead through of estimation of complication of algorithms of EKF-SLAM and construction of map of locality in accordance with supporting points, from point of his algorithmically programmatic realization. It enables to determine the ways of subsequent development and adaptation of the known mathematical correlations of algorithms of EKF-SLAM and DP-SLAM for diminishing of errors of calculations of co-ordinates airborne COMPUTERS of autonomous mobile object for realization of algorithms. The estimation of the state of off-line mobile unit is arrived at by filtration of particles. The great number of hypotheses which are an eventual number is generated, which show by itself the predictable place of location of robot. Every meaningful element of map, that orienteer, in every particle can be appraised with the use of the extended filters of Kalmana, particles of robot conditioned position. And the coefficient of weight of particles settles accounts for determination of probability of hit of certain part in a final set, which will present not only the real place of location of autonomous mobile object on a map but also position of found out all orienteers. The way of modification of the known mathematical correlations of filters of Kalmana offered in-process from point of their adaptation to the features of algorithmic and programmatic realization in airborne COMPUTERS provides economy of memory of airborne COMPUTER and diminishing of necessary calculable resource It is noticed that the algorithms of realization of SLAM of navigation are changed the offered way use less of particles, than methods, based only on a frequency filter. The error of initial calculation of co-ordinates of orienteer is taken to the minimum and does not accumulate in course of time in mathematical sense.
25
Zhang, Ye’e, and Xiaoxia Song. "A Multi-Strategy Adaptive Comprehensive Learning PSO Algorithm and Its Application." Entropy 24, no. 7 (June 28, 2022): 890. http://dx.doi.org/10.3390/e24070890.
Full textAbstract:
In this paper, a multi-strategy adaptive comprehensive learning particle swarm optimization algorithm is proposed by introducing the comprehensive learning, multi-population parallel, and parameter adaptation. In the proposed algorithm, a multi-population parallel strategy is designed to improve population diversity and accelerate convergence. The population particle exchange and mutation are realized to ensure information sharing among the particles. Then, the global optimal value is added to velocity update to design a new velocity update strategy for improving the local search ability. The comprehensive learning strategy is employed to construct learning samples, so as to effectively promote the information exchange and avoid falling into local extrema. By linearly changing the learning factors, a new factor adjustment strategy is developed to enhance the global search ability, and a new adaptive inertia weight-adjustment strategy based on an S-shaped decreasing function is developed to balance the search ability. Finally, some benchmark functions and the parameter optimization of photovoltaics are selected. The proposed algorithm obtains the best performance on 6 out of 10 functions. The results show that the proposed algorithm has greatly improved diversity, solution accuracy, and search ability compared with some variants of particle swarm optimization and other algorithms. It provides a more effective parameter combination for the complex engineering problem of photovoltaics, so as to improve the energy conversion efficiency.
26
du Plessis, Marthinus Christoffel, Hiroaki Shiino, and Masashi Sugiyama. "Online Direct Density-Ratio Estimation Applied to Inlier-Based Outlier Detection." Neural Computation 27, no. 9 (September 2015): 1899–914. http://dx.doi.org/10.1162/neco_a_00761.
Full textAbstract:
Many machine learning problems, such as nonstationarity adaptation, outlier detection, dimensionality reduction, and conditional density estimation, can be effectively solved by using the ratio of probability densities. Since the naive two-step procedure of first estimating the probability densities and then taking their ratio performs poorly, methods to directly estimate the density ratio from two sets of samples without density estimation have been extensively studied recently. However, these methods are batch algorithms that use the whole data set to estimate the density ratio, and they are inefficient in the online setup, where training samples are provided sequentially and solutions are updated incrementally without storing previous samples. In this letter, we propose two online density-ratio estimators based on the adaptive regularization of weight vectors. Through experiments on inlier-based outlier detection, we demonstrate the usefulness of the proposed methods.
27
Taimoor, Muhammad, Xiao Lu, Hamid Maqsood, and Chunyang Sheng. "Adaptive rapid neural observer-based sensors fault diagnosis and reconstruction of quadrotor unmanned aerial vehicle." Aircraft Engineering and Aerospace Technology 93, no. 5 (June 17, 2021): 847–61. http://dx.doi.org/10.1108/aeat-01-2021-0005.
Full textAbstract:
Purpose The objective of this research is to investigate various neural network (NN) observer techniques for sensors fault identification and diagnosis of nonlinear system in consideration of numerous faults, failures, uncertainties and disturbances. For the importunity of increasing the faults diagnosis and reconstruction preciseness, a new technique is used for modifying the weight parameters of NNs without enhancement of computational complexities. Design/methodology/approach Various techniques such as adaptive radial basis functions (ARBF), conventional radial basis functions, adaptive multi-layer perceptron, conventional multi-layer perceptron and extended state observer are presented. For increasing the fault detection preciseness, a new technique is used for updating the weight parameters of radial basis functions and multi-layer perceptron (MLP) without enhancement of computational complexities. Lyapunov stability theory and sliding-mode surface concepts are used for the weight-updating parameters. Based on the combination of these two concepts, the weight parameters of NNs are updated adaptively. The key purpose of utilization of adaptive weight is to enhance the detection of faults with high accuracy. Because of the online adaptation, the ARBF can detect various kinds of faults and failures such as simultaneous, incipient, intermittent and abrupt faults effectively. Results depict that the suggested algorithm (ARBF) demonstrates more confrontation to unknown disturbances, faults and system dynamics compared with other investigated techniques and techniques used in the literature. The proposed algorithms are investigated by the utilization of quadrotor unmanned aerial vehicle dynamics, which authenticate the efficiency of the suggested algorithm. Findings The proposed Lyapunov function theory and sliding-mode surface-based strategy are studied, which shows more efficiency to unknown faults, failures, uncertainties and disturbances compared with conventional approaches as well as techniques used in the literature. Practical implications For improvement of the system safety and for avoiding failure and damage, the rapid fault detection and isolation has a great significance; the proposed approaches in this research work guarantee the detection and reconstruction of unknown faults, which has a great significance for practical life. Originality/value In this research, two strategies such Lyapunov function theory and sliding-mode surface concept are used in combination for tuning the weight parameters of NNs adaptively. The main purpose of these strategies is the fault diagnosis and reconstruction with high accuracy in terms of shape as well as the magnitude of unknown faults. Results depict that the proposed strategy is more effective compared with techniques used in the literature.
28
Kumalija, Elhard James, and Yukikazu Nakamoto. "MiniatureVQNet: A Light-Weight Deep Neural Network for Non-Intrusive Evaluation of VoIP Speech Quality." Applied Sciences 13, no. 4 (February 14, 2023): 2455. http://dx.doi.org/10.3390/app13042455.
Full textAbstract:
In IP audio systems, audio quality is degraded by environmental noise, poor network quality, and encoding–decoding algorithms. Therefore, there is a need for a continuous automatic quality evaluation of the transmitted audio. Speech quality monitoring in VoIP systems enables autonomous system adaptation. Furthermore, there are diverse IP audio transmitters and receivers, from high-performance computers and mobile phones to low-memory and low-computing-capacity embedded systems. This paper proposes MiniatureVQNet, a single-ended speech quality evaluation method for VoIP audio applications based on a lightweight deep neural network (DNN) model. The proposed model can predict the audio quality independent of the source of degradation, whether noise or network, and is light enough to run in embedded systems. Two variations of the proposed MiniatureVQNet model were evaluated: a MiniatureVQNet model trained on a dataset that contains environmental noise only, referred to as MiniatureVQNet–Noise, and a second model trained on both noise and network distortions, referred to as MiniatureVQNet–Noise–Network. The proposed MiniatureVQNet model outperforms the traditional P.563 method in terms of accuracy on all tested network conditions and environmental noise parameters. The mean squared error (MSE) of the models compared to the PESQ score for ITU-T P.563, MiniatureVQNet-Noise, and MiniatureVQNet–Noise–Network was 2.19, 0.34, and 0.21, respectively. The performance of both the MiniatureVQNet–Noise–Network and MiniatureVQNet-Noise model depends on the noise type for an SNR greater than 0 dB and less than 10 dB. In addition, training on a noise–network-distorted speech dataset improves the model prediction accuracy in all VoIP environment distortions compared to training the model on a noise-only dataset.
29
Hartoko, Agus, Yoan Teresia Sembiring, and Nurul Latifah. "Seagrass Cholorophyll-a, Biomass and Carbon Algorithms Based on the Field and Sentinel-2A Satellite Data at Karimunjawa Island, Indonesia." Science and Technology Indonesia 6, no. 3 (July 22, 2021): 121–30. http://dx.doi.org/10.26554/sti.2021.6.3.121-130.
Full textAbstract:
Chlorophyll-a in seagrass biomass is functioned for the photosynthetic process and store the organic carbon in their biomass of the leaf, rhizome, and root. Ecologically has functioned as blue carbon in reducing global warming adaptation and mitigation strategy. The study aimed to explore seagrass species, chlorophyll-a content, biomass and carbon stock at Karimunjawa Island. Develop algorithms of the Sentinel-2A satellite data based on field seagrass chlorophyll-a, biomass and carbon and at Pokemon and Bobby beach Karimunjawa Island. Four species of seagrass found at Bobby and Pokemon beach are Holodule pinifolia with a density of 160.44 ind.m−2 , Enhalus acoroides with 26.22 ind.m−2, Halophila ovalis with 6.67 ind.m−2 and Thalassia hemprichii with 4.44 ind.m−2.The lowest seagrass chlorophyll-a is 5.854 mg.ml−1 found in H. pinifolia and the highest is 20.819 mg.ml−1found in E. acoroides at Pokemon beach. The range of seagrass chlorophyll-a at Bobby beach was 3.485 - 14.133 mg.ml−1 in T. hemprichii. The smallest individual biomass dry weight was found in T.hempirichii with 1.32 g.dry.weight per individu, and the biggest in E.acoroides with 6.98 g.dry.weight per individu. The highest seagrass biomass at Pokemon beach was in E. acoroides with 236.93 g.m−2 which has a wide leaf morphology and the lowest in H. pinifolia with 75.91 g.m−2 with the smallest leaf morphology. The range of seagrass biomass at Bobby beach is 97.62 - 264.48 g.m−2 which is dominated by T.hempirichii. The range of seagrass carbon was 109.63 - 136.82 gC.m−2at Pokemon beach, and in the range of 95.00 - 114.01 gC.m−2 at Bobby beach. Algorithm of seagrass chlorophyll-a = -36.308 (B3/B4)2 – 140.41(B3/B4) + 83.912 ; biomass = -7028.3 (B3/B4)2 + 14948 (B3/B4) – 7764.4; carbon = -17.529(B2/B3)2 + 143.82(B2/B3) – 5.3362 for Pokemon beach. Algorithm of chlorophyll-a = 455.02 (B2/B4)2 + 823.72 (B2/B4) + 375.48; biomass = -14699 (B3/B2)2 + 28395(B3/B2) – 13537; and carbon = - 0.001(B3/B4)2+ 0.209(B3/B4) - 10.203 for Bobby beach. The use of Band-2 (0.490 ????m), Band-3 (0.560 ????m) and Band-4 (0.665 ????m) Sentinel-2A satellite data in the development of seagras chlorophyll-a, biomass and carbon algorithm was found to be significant.
30
Hanson, Stephen José, and Michiro Negishi. "On the Emergence of Rules in Neural Networks." Neural Computation 14, no. 9 (September 1, 2002): 2245–68. http://dx.doi.org/10.1162/089976602320264079.
Full textAbstract:
A simple associationist neural network learns to factor abstract rules (i.e., grammars) from sequences of arbitrary input symbols by inventing abstract representations that accommodate unseen symbol sets as well as unseen but similar grammars. The neural network is shown to have the ability to transfer grammatical knowledge to both new symbol vocabularies and new grammars. Analysis of the state-space shows that the network learns generalized abstract structures of the input and is not simply memorizing the input strings. These representations are context sensitive, hierarchical, and based on the state variable of the finite-state machines that the neural network has learned. Generalization to new symbol sets or grammars arises from the spatial nature of the internal representations used by the network, allowing new symbol sets to be encoded close to symbol sets that have already been learned in the hidden unit space of the network. The results are counter to the arguments that learning algorithms based on weight adaptation after each exemplar presentation (such as the long term potentiation found in the mammalian nervous system) cannot in principle extract symbolic knowledge from positive examples as prescribed by prevailing human linguistic theory and evolutionary psychology.
31
Прилуцкая, В. А., А. В. Сукало, and А. В. Сапотницкий. "Clinical-Anamnestic and Anthropometric Features of Small for Gestational Age Full-Term Newborns." Педиатрия. Восточная Европа, no. 2 (June 17, 2021): 244–59. http://dx.doi.org/10.34883/pi.2021.9.2.009.
Full textAbstract:
Цель. Проанализировать особенности состояния здоровья доношенных детей, рожденных маловесными к сроку гестации.Материалы и методы. Проведено динамическое медицинское обследование 299 доношенных новорожденных, рожденных и получавших лечение в ГУ «РНПЦ «Мать и дитя». Дети разделены на 2 группы: группа 1 (Гр1) – маловесные к сроку гестации (n=160), группа контроля (Гр2) – новорожденные с соответствующим сроку гестации физическим развитием (n=139). Изучены клинико-анамнестические данные, закономерности постнатальной адаптации, заболеваний маловесных к сроку гестации младенцев в неонатальном периоде. В динамике неонатального периода проведен анализ показателей антропометрического статуса.Результаты. Выявлены более низкие средние показатели роста родителей маловесных новорожденных (матери р=0,002, отцы р=0,034) и низкие показатели прибавки массы тела женщин за время беременности (11,6±4,4 кг в Гр1 и 13,5±4,3 кг в Гр2, р<0,001). У матерей детей Гр1 значимо чаще диагностирован хронический пиелонефрит (9,38% против 1,44%, р=0,002). Доминирующими осложнениями беременности матерей маловесных детей были хроническая фетоплацентарная недостаточность (31,88% в Гр1 и 10,79% в Гр2, р<0,001), кольпит (41,88% и 29,49%, р=0,026) и анемия беременных (22,50% и 10,79%, р=0,007). У маловесных новорожденных значимо чаще отмечались нарушения постнатальной адаптации. Среди неонатальных патологических состояний статистически значимо чаще диагностированы врожденная инфекция неуточненная (р<0,001), церебральная ишемия (р<0,001), синдром угнетения ЦНС (р=0,001), синдром дезадаптации сердечно-сосудистой системы (р=0,014) и малые аномалии развития сердца (р=0,041). У обследованных новорожденных не наблюдалось скачков в росте. Показатели физического развития маловесных новорожденных не имели различий при делении по половому признаку. Антропометрические показатели маловесных детей в динамике неонатального периода сохранялись статистически значимо более низкими, чем у младенцев контрольной группы, при этом средние прибавки массы за первый месяц превысили нормативные значения.Выводы. Выявленные клинические, анамнестические и антропометрические особенности у маловесных новорожденных объясняют высокую частоту нарушений адаптации иповышенный риск заболеваний в неонатальном периоде. Установленные закономерности обосновывают важность тщательного неонатального мониторинга, разработки алгоритмов оказания медицинской помощи и персонифицированного диспансерного наблюдения доношенных детей, рожденных маловесными к сроку гестации. Purpose. To analyze the features of the health status of full-term small for gestational age infants. Materials and methods. A dynamic medical examination of 299 full-term newborns, born and treated at the State Institution "Republican Scientific and Practical Center "Mother and Child", was carried out. Children were divided into 2 groups: group 1 (Gr1) – small for gestational age (n=160) and control group (Gr2) – newborns with physical development corresponding to gestational age (n=139). The clinical and anamnestic data, regularities of postnatal adaptation, diseases of low birth weight infants in the neonatal period were studied. In the dynamics of the neonatal period, the analysis of the indicators of anthropometric status was carried out.Results. Lower average growth rates of parents of low birth weight infants (mothers p=0.002, fathers p=0.034) and low rates of weight gain of a woman during pregnancy (11.6±4.4 kg in Gr1 and 13.5±4.3 kg in Gr2, p<0.001) were revealed. Chronic pyelonephritis was diagnosed significantly more often in mothers of the Gr1 children (9.38% versus 1.44%, p=0.002). The dominant complications of pregnancy in low birth weight mothers were chronic placental insufficiency (31.88% in Gr1 and 10.79% in Gr2, p<0.001), colpitis (41.88% and 29.49%, p=0.026), and anemia of pregnant women (22.50% and 10.79%, p=0.007). Small for gestational age infants had postnatal adaptation disorders significantly more frequently. Among neonatal pathological conditions, congenital unspecified infection (p<0.001), cerebral ischemia (p<0.001), central nervous system depression syndrome (p=0.001), maladjustment syndrome of the cardiovascular system (p=0.014), and minor anomalies of the heart (p=0.041) were diagnosed statistically significantly more often. The surveyed newborns showed no growth spurt. The indices of physical development of low birth weight infants did not differ when divided by sex. The anthropometric indicators of small for gestational age children in the dynamics of the neonatal period remained statistically significantly lower than in infants of the control group, while the average weight gain for the first month exceeded the standard values.Conclusions. The revealed clinical, anamnestic and anthropometric features in small for gestational age newborns explain the high frequency of adaptation disorders and the increased risk of diseases in the neonatal period. The established patterns substantiate the importance of careful neonatal monitoring, development of algorithms for provision of medical care and personalized dispensary observation of full-term small for gestational age babies.
32
Nicol, Ginger E., Amanda R. Ricchio, Christopher L. Metts, Michael D. Yingling, Alex T. Ramsey, Julia A. Schweiger, J. Philip Miller, and Eric J. Lenze. "A Smartphone-Based Technique to Detect Dynamic User Preferences for Tailoring Behavioral Interventions: Observational Utility Study of Ecological Daily Needs Assessment." JMIR mHealth and uHealth 8, no. 11 (November 13, 2020): e18609. http://dx.doi.org/10.2196/18609.
Full textAbstract:
Background Mobile health apps are promising vehicles for delivering scalable health behavior change interventions to populations that are otherwise difficult to reach and engage, such as young adults with psychiatric conditions. To improve uptake and sustain consumer engagement, mobile health interventions need to be responsive to individuals’ needs and preferences, which may change over time. We previously created an ecological daily needs assessment to capture microprocesses influencing user needs and preferences for mobile health treatment adaptation. Objective The objective of our study was to test the utility of a needs assessment anchored within a mobile app to capture individualized, contextually relevant user needs and preferences within the framework of a weight management mobile health app. Methods Participants with an iOS device could download the study app via the study website or links from social media. In this fully remote study, we screened, obtained informed consent from, and enrolled participants through the mobile app. The mobile health framework included daily health goal setting and self-monitoring, with up to 6 daily prompts to determine in-the-moment needs and preferences for mobile health–assisted health behavior change. Results A total of 24 participants downloaded the app and provided e-consent (22 female; 2 male), with 23 participants responding to at least one prompt over 2 weeks. The mean length of engagement was 5.6 (SD 4.7) days, with a mean of 2.8 (1.1) responses per day. We observed individually dynamic needs and preferences, illustrating daily variability within and between individuals. Qualitative feedback indicated preferences for self-adapting features, simplified self-monitoring, and the ability to personalize app-generated message timing and content. Conclusions The technique provided an individually dynamic and contextually relevant alternative and complement to traditional needs assessment for assessing individually dynamic user needs and preferences during treatment development or adaptation. The results of this utility study suggest the importance of personalization and learning algorithms for sustaining app engagement in young adults with psychiatric conditions. Further study in broader user populations is needed.
33
Fiori, Simone. "Nonlinear Complex-Valued Extensions of Hebbian Learning: An Essay." Neural Computation 17, no. 4 (April 1, 2005): 779–838. http://dx.doi.org/10.1162/0899766053429381.
Full textAbstract:
The Hebbian paradigm is perhaps the best-known unsupervised learning theory in connectionism. It has inspired wide research activity in the artificial neural network field because it embodies some interesting properties such as locality and the capability of being applicable to the basic weight-and-sum structure of neuron models. The plain Hebbian principle, however, also presents some inherent theoretical limitations that make it impractical in most cases. Therefore, modifications of the basic Hebbian learning paradigm have been proposed over the past 20 years in order to design profitable signal and data processing algorithms. Such modifications led to the principal component analysis type class of learning rules along with their nonlinear extensions. The aim of this review is primarily to present part of the existing fragmented material in the field of principal component learning within a unified view and contextually to motivate and present extensions of previous works on Hebbian learning to complex-weighted linear neural networks. This work benefits from previous studies on linear signal decomposition by artificial neural networks, nonquadratic component optimization and reconstruction error definition, neural parameters adaptation by constrained optimization of learning criteria of complex-valued arguments, and orthonormality expression via the insertion of topological elements in the networks or by modifying the network learning criterion. In particular, the learning principles considered here and their analysis concern complex-valued principal/minor component/subspace linear/nonlinear rules for complex-weighted neural structures, both feedforward and laterally connected.
34
SUGANO, Youhei, and Hiroshi IGARASHI. "Environment Adaptation Algorithm using Personality Imitation and Imitation Weight." Proceedings of the Conference on Information, Intelligence and Precision Equipment : IIP 2020 (2020): P14. http://dx.doi.org/10.1299/jsmeiip.2020.p14.
Full text
35
Ardekani, Iman Tabatabaei, and Waleed H. Abdulla. "Filtered weight FxLMS adaptation algorithm: Analysis, design and implementation." International Journal of Adaptive Control and Signal Processing 25, no. 11 (June 1, 2011): 1023–37. http://dx.doi.org/10.1002/acs.1257.
Full text
36
Qi, Jin, and Jie Hu. "Multivariable case-based reason adaptation based on multiple-output support vector regression with similarity-related weight for parametric mechanical design." Advances in Mechanical Engineering 10, no. 10 (October 2018): 168781401880464. http://dx.doi.org/10.1177/1687814018804649.
Full textAbstract:
Using historical cases’ solutions to obtain feasible solution for new problem is fundamentally to successfully applying case-based reason technique in parametric mechanical design. As a well-known intelligent algorithm, the formulation of support vector regression has been taken for case-based reason adaptation, but the standard support vector regression can only be used as a univariate adaptation method because of its single-output structure, which would result in the ignorance of the possible interrelations among solution outputs. To handle the complicated case adaptation task with large number of problem inputs and solution outputs more efficiently, this study investigates the possibility of multivariable case-based reason adaptation with multiple output by applying multiple-output support vector regression. Furthermore, inspired by the fact that training sample which contains two closer cases can provide more useful information than others, this study adds the similarity-related weight into multiple-output support vector regression and gives high weights to the information provided by such useful training sample during multi-dimensional regression estimation. The superiority of proposed multiple-output support vector regression with similarity-related weight is validated by the actual design example and quantitative comparisons with other adaptation methods. The comparative results indicate that multiple-output support vector regression with similarity-related weight achieves the best performance for large-quantity case-based reason adaptation because of its higher accuracy and relatively lower cost.
37
Gao, Peng, Jingmei Li, Guodong Zhao, and Changhong Ding. "Multisource Deep Transfer Learning Based on Balanced Distribution Adaptation." Computational Intelligence and Neuroscience 2022 (April 18, 2022): 1–12. http://dx.doi.org/10.1155/2022/6915216.
Full textAbstract:
The current traditional unsupervised transfer learning assumes that the sample is collected from a single domain. From the aspect of practical application, the sample from a single-source domain is often not enough. In most cases, we usually collect labeled data from multiple domains. In recent years, multisource unsupervised transfer learning with deep learning has focused on aligning in the common feature space and then seeking to minimize the distribution difference between the source and target domains, such as marginal distribution, conditional distribution, or both. Moreover, conditional distribution and marginal distribution are often treated equally, which will lead to poor performance in practical applications. The existing algorithms that consider balanced distribution are often based on a single-source domain. To solve the above-mentioned problems, we propose a multisource transfer learning algorithm based on distribution adaptation. This algorithm considers adjusting the weights of two distributions to solve the problem of distribution adaptation in multisource transfer learning. A large number of experiments have shown that our method MTLBDA has achieved significant results in popular image classification datasets such as Office-31.
38
HUANG, Fang, Zhijun FANG, Zhicai SHI, Lehui ZHUANG, Xingchen LI, and Bo HUANG. "A Federated Domain Adaptation Algorithm Based on Knowledge Distillation and Contrastive Learning." Wuhan University Journal of Natural Sciences 27, no. 6 (December 2022): 499–507. http://dx.doi.org/10.1051/wujns/2022276499.
Full textAbstract:
Smart manufacturing suffers from the heterogeneity of local data distribution across parties, mutual information silos and lack of privacy protection in the process of industry chain collaboration. To address these problems, we propose a federated domain adaptation algorithm based on knowledge distillation and contrastive learning. Knowledge distillation is used to extract transferable integration knowledge from the different source domains and the quality of the extracted integration knowledge is used to assign reasonable weights to each source domain. A more rational weighted average aggregation is used in the aggregation phase of the center server to optimize the global model, while the local model of the source domain is trained with the help of contrastive learning to constrain the local model optimum towards the global model optimum, mitigating the inherent heterogeneity between local data. Our experiments are conducted on the largest domain adaptation dataset, and the results show that compared with other traditional federated domain adaptation algorithms, the algorithm we proposed trains a more accurate model, requires fewer communication rounds, makes more effective use of imbalanced data in the industrial area, and protects data privacy.
39
Banister, B. C., and J. R. Zeidler. "A simple gradient sign algorithm for transmit antenna weight adaptation with feedback." IEEE Transactions on Signal Processing 51, no. 5 (May 2003): 1156–71. http://dx.doi.org/10.1109/tsp.2002.808104.
Full text
40
Chin, T. M., M. J. Turmon, J. B. Jewell, and M. Ghil. "An Ensemble-Based Smoother with Retrospectively Updated Weights for Highly Nonlinear Systems." Monthly Weather Review 135, no. 1 (January 1, 2007): 186–202. http://dx.doi.org/10.1175/mwr3353.1.
Full textAbstract:
Abstract Monte Carlo computational methods have been introduced into data assimilation for nonlinear systems in order to alleviate the computational burden of updating and propagating the full probability distribution. By propagating an ensemble of representative states, algorithms like the ensemble Kalman filter (EnKF) and the resampled particle filter (RPF) rely on the existing modeling infrastructure to approximate the distribution based on the evolution of this ensemble. This work presents an ensemble-based smoother that is applicable to the Monte Carlo filtering schemes like EnKF and RPF. At the minor cost of retrospectively updating a set of weights for ensemble members, this smoother has demonstrated superior capabilities in state tracking for two highly nonlinear problems: the double-well potential and trivariate Lorenz systems. The algorithm does not require retrospective adaptation of the ensemble members themselves, and it is thus suited to a streaming operational mode. The accuracy of the proposed backward-update scheme in estimating non-Gaussian distributions is evaluated by comparison to the more accurate estimates provided by a Markov chain Monte Carlo algorithm.
41
Abbas, Nizar Hadi. "Tuning of different controlling techniques for magnetic suspending system using an improved bat algorithm." International Journal of Electrical and Computer Engineering (IJECE) 10, no. 3 (June 1, 2020): 2402. http://dx.doi.org/10.11591/ijece.v10i3.pp2402-2415.
Full textAbstract:
In this paper, design of proportional- derivative (PD) controller, pseudo-derivative-feedback (PDF) controller and PDF with feedforward (PDFF) controller for magnetic suspending system have been presented. Tuning of the above controllers is achieved based on Bat algorithm (BA). BA is a recent bio-inspired optimization method for solving global optimization problems, which mimic the behavior of micro-bats. The weak point of the standard BA is the exploration ability due to directional echolocation and the difficulty in escaping from local optimum. The new improved BA enhances the convergence rate while obtaining optimal solution by introducing three adaptations namely modified frequency factor, adding inertia weight and modified local search. The feasibility of the proposed algorithm is examined by applied to several benchmark problems that are adopted from literature. The results of IBA are compared with the results collected from standard BA and the well-known particle swarm optimization (PSO) algorithm. The simulation results show that the IBA has a higher accuracy and searching speed than the approaches considered. Finally, the tuning of the three controlling schemes using the proposed algorithm, standard BA and PSO algorithms reveals that IBA has a higher performance compared with the other optimization algorithms
42
Yan, Wei, Qi Gao, Zheng Gang Liu, Shan Hui Zhang, and Yu Ping Hu. "A Novel Discovery Technique on Attribute Weight of Engine CBR Design System." Advanced Materials Research 97-101 (March 2010): 3714–17. http://dx.doi.org/10.4028/www.scientific.net/amr.97-101.3714.
Full textAbstract:
An improved multi-group self-adaptive evolutionary programming Algorithm is used to get adapt attribute weight for CBR system. Firstly, this paper analyses the adaptability function based on reference case base REF and testing case base TEST, develops a novel Bi-group self-adaptive evolutionary programming that overcome the lack of conventional evolutionary programming. In this Novel algorithm, evolution of Cauchy operator and Gauss operator are parallel performed with different mutation strategies, and the Gauss operator owns the ability of self-adaptation according to the variation of adaptability function. Information is exchanged when sub-groups are reorganized. Experiment results prove the validity of self-adaptive Algorithm and CBR design system is used successfully in engine design process.
43
Sasaki, Takahiro, and Mario Tokoro. "Evolving Learnable Neural Networks Under Changing Environments with Various Rates of Inheritance of Acquired Characters: Comparison of Darwinian and Lamarckian Evolution." Artificial Life 5, no. 3 (July 1999): 203–23. http://dx.doi.org/10.1162/106454699568746.
Full textAbstract:
The processes of adaptation in natural organisms consist of two complementary phases: learning, occurring within each individual's lifetime, and evolution, occurring over successive generations of the population. In this article, we study the relationship between learning and evolution in a simple abstract model, where neural networks capable of learning are evolved using genetic algorithms (GAs). Individuals try to maximize their life energy by learning certain rules that distinguish between two groups of materials: food and poison. The connective weights of individuals' neural networks undergo modification, that is, certain characters will be acquired, through their lifetime learning. By setting various rates for the heritability of acquired characters, which is a motive force of Lamarckian evolution, we observe adaptational processes of populations over successive generations. Paying particular attention to behaviors under changing environments, we show the following results. Populations with lower rates of heritability not only show more stable behavior against environmental changes, but also maintain greater adaptability with respect to such changing environments. Consequently, the population with zero heritability, that is, the Darwinian population, attains the highest level of adaptation to dynamic environments.
44
Seregi, Bálint Leon, and Péter Ficzere. "Weight Reduction of a Drone Using Generative Design." Hungarian Journal of Industry and Chemistry 49, no. 2 (2021): 19–22. http://dx.doi.org/10.33927/hjic-2021-16.
Full textAbstract:
Generative design has the potential to be optimized with different parameters using a design method based on artificial intelligence and by defining the design problem. The use of this method on a drone frame is presented with explanations of the various design phases. The goal of the optimisation was to be able to fit a battery with a larger capacity onto the Unmanned Aerial Vehicle and compensate for its increased weight by reducing the weight of the drone frame using a generative algorithm. As the production possibilities were limited, adaptation to the selected manufacturing technology was also taken into account during the optimization.
45
Ziani, M. "Towards adaptation of the NURBS weights in shape optimization." Mathematical Modeling and Computing 9, no. 4 (2022): 959–67. http://dx.doi.org/10.23939/mmc2022.04.959.
Full textAbstract:
Bézier based parametrisations in shape optimization have the drawback of using high degree polynomials to draw more complex shapes. To overcome this drawback, Non-Uniform Rational B-Splines (NURBS) are usually used. But, by considering the NURBS weights, in addition to the locations of the control points, as optimization variables, the dimension of the problem greatly increases and this would make the optimization process stiffer. In this work we propose, then, an algorithm to adapt the weights of NURBS in the parametrization of shape optimization problems. Unlike the coordinates of the control points, the weights are not considered, in this case, as variables of the optimization process. From the knowledge of an approximate optimal shape, we consider the set of all NURBS parametrizations of the same degree that approximate the shape in the sense of least squares. Then, we elect the parametrization associated with the most regular control polygon (least length of the control polygon). Numerical results show that the adaptive parametrization improves the performance of the optimization process.
46
Fujimoto, Yu, Hideitsu Hino, and Noboru Murata. "An Estimation of Generalized Bradley-Terry Models Based on the em Algorithm." Neural Computation 23, no. 6 (June 2011): 1623–59. http://dx.doi.org/10.1162/neco_a_00129.
Full textAbstract:
The Bradley-Terry model is a statistical representation for one's preference or ranking data by using pairwise comparison results of items. For estimation of the model, several methods based on the sum of weighted Kullback-Leibler divergences have been proposed from various contexts. The purpose of this letter is to interpret an estimation mechanism of the Bradley-Terry model from the viewpoint of flatness, a fundamental notion used in information geometry. Based on this point of view, a new estimation method is proposed on a framework of the em algorithm. The proposed method is different in its objective function from that of conventional methods, especially in treating unobserved comparisons, and it is consistently interpreted in a probability simplex. An estimation method with weight adaptation is also proposed from a viewpoint of the sensitivity. Experimental results show that the proposed method works appropriately, and weight adaptation improves accuracy of the estimate.
47
Merayo, Noemí, Ramón J. Durán, Patricia Fernández, Rubén M. Lorenzo, Ignacio de Miguel, and Evaristo J. Abril. "EPON bandwidth allocation algorithm based on automatic weight adaptation to provide client and service differentiation." Photonic Network Communications 17, no. 2 (August 27, 2008): 119–28. http://dx.doi.org/10.1007/s11107-008-0147-9.
Full text
48
Burghardt, Andrzej, and Marcin Szuster. "Neuro-Dynamic Programming in Control of the Ball and Beam System." Solid State Phenomena 210 (October 2013): 206–14. http://dx.doi.org/10.4028/www.scientific.net/ssp.210.206.
Full textAbstract:
This paper presents a new approach to the control problem of the ball and beam system, with a Neuro-Dynamic Programming algorithm implemented as the main part of the control system. The controlled system is included in the group of underactuated systems, which are nonlinear dynamical objects with the number of control signals smaller than the number of degrees of freedom. This results in problems in the formulation of a stable control algorithm, that guarantees stabilization of the ball in the desired position on the beam. The type of ball and beam material has a noticeable influence on the difficulties in stabilization of the ball, because of a smaller rolling friction and big inertia of the used metallic ball in comparison to other, for example made of non-metallic materials. The main part of the proposed discrete control system is the Neuro-Dynamic Programming algorithm in a Dual-Heuristic Dynamic Programming configuration, realized in a form of two neural networks: the actor and the critic. Neuro-Dynamic Programming algorithms use the Reinforcement Learning idea for adaptation of artificial neural network weights. Additional elements of the control system are the PD controller and the supervisory term, that ensures stability of the closed system loop. The control algorithm works on-line and does not require a preliminary learning phase of the neural network weights. Performance of the control algorithm was verified using the physical system controlled by the dSpace digital signal processing board.
49
Yuan, Chunhua, Jiang Wang, and Guosheng Yi. "Estimation of key parameters in adaptive neuron model according to firing patterns based on improved particle swarm optimization algorithm." Modern Physics Letters B 31, no. 07 (March 10, 2017): 1750060. http://dx.doi.org/10.1142/s0217984917500609.
Full textAbstract:
Estimation of ion channel parameters is crucial to spike initiation of neurons. The biophysical neuron models have numerous ion channel parameters, but only a few of them play key roles in the firing patterns of the models. So we choose three parameters featuring the adaptation in the Ermentrout neuron model to be estimated. However, the traditional particle swarm optimization (PSO) algorithm is still easy to fall into local optimum and has the premature convergence phenomenon in the study of some problems. In this paper, we propose an improved method that uses a concave function and dynamic logistic chaotic mapping mixed to adjust the inertia weights of the fitness value, effectively improve the global convergence ability of the algorithm. The perfect predicting firing trajectories of the rebuilt model using the estimated parameters prove that only estimating a few important ion channel parameters can establish the model well and the proposed algorithm is effective. Estimations using two classic PSO algorithms are also compared to the improved PSO to verify that the algorithm proposed in this paper can avoid local optimum and quickly converge to the optimal value. The results provide important theoretical foundations for building biologically realistic neuron models.
50
Abu Doush, Iyad, Amal Lutfi Quran, Mohammed Azmi Al-Betar, and Mohammed A. Awadallah. "MAX-SAT Problem using Hybrid Harmony Search Algorithm." Journal of Intelligent Systems 27, no. 4 (October 25, 2018): 643–58. http://dx.doi.org/10.1515/jisys-2016-0129.
Full textAbstract:
Abstract Maximum Satisfiability problem is an optimization variant of the Satisfiability problem (SAT) denoted as MAX-SAT. The aim of this problem is to find Boolean variable assignment that maximizes the number of satisfied clauses in the Boolean formula. In case the number of variables per clause is equal or greater than three, then this problem is considered NP-complete. Hence, many researchers have developed techniques to deal with MAX-SAT. In this paper, we investigate the impact of different hybrid versions of binary harmony search (HS) algorithm on solving MAX 3-SAT problem. Therefore, we propose two novel hybrid binary HS algorithms. The first hybridizes Flip heuristic with HS, and the second uses Tabu search combined with Flip heuristic. Furthermore, a distinguished feature of our proposed approaches is using an objective function that is updated dynamically based on the stepwise adaptation of weights (SAW) mechanism to evaluate the MAX-SAT solution using the proposed hybrid versions. The performance of the proposed approaches is evaluated over standard MAX-SAT benchmarks, and the results are compared with six evolutionary algorithms and three stochastic local search algorithms. The obtained results are competitive and show that the proposed novel approaches are effective.