Journal articles: 'Linear perceptrons'

1

Bylander, Tom. "Learning Linear Threshold Approximations Using Perceptrons." Neural Computation 7, no. 2 (March 1995): 370–79. http://dx.doi.org/10.1162/neco.1995.7.2.370.

Full text

Abstract:

We demonstrate sufficient conditions for polynomial learnability of suboptimal linear threshold functions using perceptrons. The central result is as follows. Suppose there exists a vector w*, of n weights (including the threshold) with “accuracy” 1 − α, “average error” η, and “balancing separation” σ, i.e., with probability 1 − α, w* correctly classifies an example x; over examples incorrectly classified by w*, the expected value of |w* · x| is η (source of inaccuracy does not matter); and over a certain portion of correctly classified examples, the expected value of |w* · x| is σ. Then, with probability 1 − δ, the perceptron achieves accuracy at least 1 − [∊ + α(1 + η/σ)] after O[n∊−2σ−2(ln 1/δ)] examples.

APA, Harvard, Vancouver, ISO, and other styles

2

Alpaydin, E., and M. I. Jordan. "Local linear perceptrons for classification." IEEE Transactions on Neural Networks 7, no. 3 (May 1996): 788–94. http://dx.doi.org/10.1109/72.501737.

Full text

APA, Harvard, Vancouver, ISO, and other styles

3

Barber, D., D. Saad, and P. Sollich. "Test Error Fluctuations in Finite Linear Perceptrons." Neural Computation 7, no. 4 (July 1995): 809–21. http://dx.doi.org/10.1162/neco.1995.7.4.809.

Full text

Abstract:

We examine the fluctuations in the test error induced by random, finite, training and test sets for the linear perceptron of input dimension n with a spherically constrained weight vector. This variance enables us to address such issues as the partitioning of a data set into a test and training set. We find that the optimal assignment of the test set size scales with n2/3.

APA, Harvard, Vancouver, ISO, and other styles

4

Legenstein, Robert, and Wolfgang Maass. "On the Classification Capability of Sign-Constrained Perceptrons." Neural Computation 20, no. 1 (January 2008): 288–309. http://dx.doi.org/10.1162/neco.2008.20.1.288.

Full text

Abstract:

The perceptron (also referred to as McCulloch-Pitts neuron, or linear threshold gate) is commonly used as a simplified model for the discrimination and learning capability of a biological neuron. Criteria that tell us when a perceptron can implement (or learn to implement) all possible dichotomies over a given set of input patterns are well known, but only for the idealized case, where one assumes that the sign of a synaptic weight can be switched during learning. We present in this letter an analysis of the classification capability of the biologically more realistic model of a sign-constrained perceptron, where the signs of synaptic weights remain fixed during learning (which is the case for most types of biological synapses). In particular, the VC-dimension of sign-constrained perceptrons is determined, and a necessary and sufficient criterion is provided that tells us when all 2m dichotomies over a given set of m patterns can be learned by a sign-constrained perceptron. We also show that uniformity of L1 norms of input patterns is a sufficient condition for full representation power in the case where all weights are required to be nonnegative. Finally, we exhibit cases where the sign constraint of a perceptron drastically reduces its classification capability. Our theoretical analysis is complemented by computer simulations, which demonstrate in particular that sparse input patterns improve the classification capability of sign-constrained perceptrons.

APA, Harvard, Vancouver, ISO, and other styles

5

Yu, Xin, Mian Xie, Li Xia Tang, and Chen Yu Li. "Learning Algorithm for Fuzzy Perceptron with Max-Product Composition." Applied Mechanics and Materials 687-691 (November 2014): 1359–62. http://dx.doi.org/10.4028/www.scientific.net/amm.687-691.1359.

Full text

Abstract:

Fuzzy neural networks is a powerful computational model, which integrates fuzzy systems with neural networks, and fuzzy perceptron is a kind of this neural networks. In this paper, a learning algorithm is proposed for a fuzzy perceptron with max-product composition, and the topological structure of this fuzzy perceptron is the same as conventional linear perceptrons. The inner operations involved in the working process of this fuzzy perceptron are based on the max-product logical operations rather than conventional multiplication and summation etc. To illustrate the finite convergence of proposed algorithm, some numerical experiments are provided.

APA, Harvard, Vancouver, ISO, and other styles

6

Shah, J. V., and Chi-Sang Poon. "Linear independence of internal representations in multilayer perceptrons." IEEE Transactions on Neural Networks 10, no. 1 (1999): 10–18. http://dx.doi.org/10.1109/72.737489.

Full text

APA, Harvard, Vancouver, ISO, and other styles

7

Zwietering, P. J., E. H. L. Aarts, and J. Wessels. "EXACT CLASSIFICATION WITH TWO-LAYERED PERCEPTRONS." International Journal of Neural Systems 03, no. 02 (January 1992): 143–56. http://dx.doi.org/10.1142/s0129065792000127.

Full text

Abstract:

We study the capabilities of two-layered perceptrons for classifying exactly a given subset. Both necessary and sufficient conditions are derived for subsets to be exactly classifiable with two-layered perceptrons that use the hard-limiting response function. The necessary conditions can be viewed as generalizations of the linear-separability condition of one-layered perceptrons and confirm the conjecture that the capabilities of two-layered perceptrons are more limited than those of three-layered perceptrons. The sufficient conditions show that the capabilities of two-layered perceptrons extend beyond the exact classification of convex subsets. Furthermore, we present an algorithmic approach to the problem of verifying the sufficiency condition for a given subset.

APA, Harvard, Vancouver, ISO, and other styles

8

Hara, Kazuyuki, and Masato Okada. "Ensemble Learning of Linear Perceptrons: On-Line Learning Theory." Journal of the Physical Society of Japan 74, no. 11 (November 15, 2005): 2966–72. http://dx.doi.org/10.1143/jpsj.74.2966.

Full text

APA, Harvard, Vancouver, ISO, and other styles

9

Frean, Marcus. "The Upstart Algorithm: A Method for Constructing and Training Feedforward Neural Networks." Neural Computation 2, no. 2 (June 1990): 198–209. http://dx.doi.org/10.1162/neco.1990.2.2.198.

Full text

Abstract:

A general method for building and training multilayer perceptrons composed of linear threshold units is proposed. A simple recursive rule is used to build the structure of the network by adding units as they are needed, while a modified perceptron algorithm is used to learn the connection strengths. Convergence to zero errors is guaranteed for any boolean classification on patterns of binary variables. Simulations suggest that this method is efficient in terms of the numbers of units constructed, and the networks it builds can generalize over patterns not in the training set.

APA, Harvard, Vancouver, ISO, and other styles

10

Hamid, Danish, Syed Sajid Ullah, Jawaid Iqbal, Saddam Hussain, Ch Anwar ul Hassan, and Fazlullah Umar. "A Machine Learning in Binary and Multiclassification Results on Imbalanced Heart Disease Data Stream." Journal of Sensors 2022 (September 20, 2022): 1–13. http://dx.doi.org/10.1155/2022/8400622.

Full text

Abstract:

In medical filed, predicting the occurrence of heart diseases is a significant piece of work. Millions of healthcare-related complexities that have remained unsolved up until now can be greatly simplified with the help of machine learning. The proposed study is concerned with the cardiac disease diagnosis decision support system. An OpenML repository data stream with 1 million instances of heart disease and 14 features is used for this study. After applying to preprocess and feature engineering techniques, machine learning approaches like random forest, decision trees, gradient boosted trees, linear support vector classifier, logistic regression, one-vs-rest, and multilayer perceptron are used to perform binary and multiclassification on the data stream. When combined with the Max Abs Scaler technique, the multilayer perceptron performed satisfactorily in both binary (Accuracy 94.8%) and multiclassification (accuracy 88.2%). Compared to the other binary classification algorithms, the GBT delivered the right outcome (accuracy of 95.8%). Multilayer perceptrons, however, did well in multiple classifications. Techniques such as oversampling and undersampling have a negative impact on disease prediction. Machine learning methods like multilayer perceptrons and ensembles can be helpful for diagnosing cardiac conditions. For this kind of unbalanced data stream, sampling techniques like oversampling and undersampling are not practical.

APA, Harvard, Vancouver, ISO, and other styles

11

Banda, Peter, Christof Teuscher, and Matthew R. Lakin. "Online Learning in a Chemical Perceptron." Artificial Life 19, no. 2 (April 2013): 195–219. http://dx.doi.org/10.1162/artl_a_00105.

Full text

Abstract:

Autonomous learning implemented purely by means of a synthetic chemical system has not been previously realized. Learning promotes reusability and minimizes the system design to simple input-output specification. In this article we introduce a chemical perceptron, the first full-featured implementation of a perceptron in an artificial (simulated) chemistry. A perceptron is the simplest system capable of learning, inspired by the functioning of a biological neuron. Our artificial chemistry is deterministic and discrete-time, and follows Michaelis-Menten kinetics. We present two models, the weight-loop perceptron and the weight-race perceptron, which represent two possible strategies for a chemical implementation of linear integration and threshold. Both chemical perceptrons can successfully identify all 14 linearly separable two-input logic functions and maintain high robustness against rate-constant perturbations. We suggest that DNA strand displacement could, in principle, provide an implementation substrate for our model, allowing the chemical perceptron to perform reusable, programmable, and adaptable wet biochemical computing.

APA, Harvard, Vancouver, ISO, and other styles

12

Sollich, P. "Finite-size effects in learning and generalization in linear perceptrons." Journal of Physics A: Mathematical and General 27, no. 23 (December 7, 1994): 7771–84. http://dx.doi.org/10.1088/0305-4470/27/23/020.

Full text

APA, Harvard, Vancouver, ISO, and other styles

13

Katahira, Kentaro, Tatsuya Cho, Kazuo Okanoya, and Masato Okada. "Optimal node perturbation in linear perceptrons with uncertain eligibility trace." Neural Networks 23, no. 2 (March 2010): 219–25. http://dx.doi.org/10.1016/j.neunet.2009.11.013.

Full text

APA, Harvard, Vancouver, ISO, and other styles

14

Chen, S., G. J. Gibson, C. F. N. Cowan, and P. M. Grant. "Adaptive equalization of finite non-linear channels using multilayer perceptrons." Signal Processing 20, no. 2 (June 1990): 107–19. http://dx.doi.org/10.1016/0165-1684(90)90122-f.

Full text

APA, Harvard, Vancouver, ISO, and other styles

15

Velásquez, Juan David, Fernán Alonso Villa, and Reinaldo C. Souza. "Time series forecasting using cascade correlation networks." Ingeniería e Investigación 30, no. 1 (January 1, 2010): 157–62. http://dx.doi.org/10.15446/ing.investig.v30n1.15226.

Full text

Abstract:

Artificial neural networks, especially multilayer perceptrons, have been recognised as being a powerful technique for forecasting nonlinear time series; however, cascade-correlation architecture is a strong competitor in this task due to it incorporating several advantages related to the statistical identification of multilayer perceptrons. This paper compares the accuracy of a cascade-correlation neural network to the linear approach, multilayer perceptrons and dynamic architecture for artificial neural networks (DAN2) to determine whether the cascade-correlation network was able to forecast the time series being studied with more accuracy. It was concluded that cascade-correlation was able to forecast time series with more accuracy than other approaches.

APA, Harvard, Vancouver, ISO, and other styles

16

Zheng, Chun-Hou, De-Shuang Huang, Kang Li, George Irwin, and Zhan-Li Sun. "MISEP Method for Postnonlinear Blind Source Separation." Neural Computation 19, no. 9 (September 2007): 2557–78. http://dx.doi.org/10.1162/neco.2007.19.9.2557.

Full text

Abstract:

In this letter, a standard postnonlinear blind source separation algorithm is proposed, based on the MISEP method, which is widely used in linear and nonlinear independent component analysis. To best suit a wide class of postnonlinear mixtures, we adapt the MISEP method to incorporate a priori information of the mixtures. In particular, a group of three-layered perceptrons and a linear network are used as the unmixing system to separate sources in the postnonlinear mixtures, and another group of three-layered perceptron is used as the auxiliary network. The learning algorithm for the unmixing system is then obtained by maximizing the output entropy of the auxiliary network. The proposed method is applied to postnonlinear blind source separation of both simulation signals and real speech signals, and the experimental results demonstrate its effectiveness and efficiency in comparison with existing methods.

APA, Harvard, Vancouver, ISO, and other styles

17

Dawson, Michael R. W., and Maya Gupta. "Probability matching in perceptrons: Effects of conditional dependence and linear nonseparability." PLOS ONE 12, no. 2 (February 17, 2017): e0172431. http://dx.doi.org/10.1371/journal.pone.0172431.

Full text

APA, Harvard, Vancouver, ISO, and other styles

18

Hara, Kazuyuki, Yoichi Nakayama, Seiji Miyoshi, and Masato Okada. "Statistical Mechanics of On-Line Mutual Learning with Many Linear Perceptrons." Journal of the Physical Society of Japan 78, no. 11 (November 15, 2009): 114001. http://dx.doi.org/10.1143/jpsj.78.114001.

Full text

APA, Harvard, Vancouver, ISO, and other styles

19

Geibel, Peter, and Fritz Wysotzki. "Learning Perceptrons and Piecewise Linear Classifiers Sensitive to Example Dependent Costs." Applied Intelligence 21, no. 1 (July 2004): 45–56. http://dx.doi.org/10.1023/b:apin.0000027766.72235.bc.

Full text

APA, Harvard, Vancouver, ISO, and other styles

20

Barber, D., D. Saad, and P. Sollich. "Finite-size effects and optimal test set size in linear perceptrons." Journal of Physics A: Mathematical and General 28, no. 5 (March 7, 1995): 1325–34. http://dx.doi.org/10.1088/0305-4470/28/5/018.

Full text

APA, Harvard, Vancouver, ISO, and other styles

21

Baklagin, Nikolaevich Vyacheslav. "Selection of Parameters and Architecture of Multilayer Perceptrons for Predicting Ice Coverage of Lakes." Ekológia (Bratislava) 36, no. 3 (September 1, 2017): 226–34. http://dx.doi.org/10.1515/eko-2017-0019.

Full text

Abstract:

Abstract The ice cover on lakes is one of the most influential factors in the lakes’ winter aquatic ecosystem. The paper presents a method for predicting ice coverage of lakes by means of multilayer perceptrons. This approach is based on historical data on the ice coverage of lakes taking Lake Onega as an example. The daily time series of ice coverage of Lake Onega for 2004–2017 was collected by means of satellite data analysis of snow and ice cover of the Northern Hemisphere. Input signals parameters for the multilayer perceptrons aimed at predicting ice coverage of lakes are based on the correlation analysis of this time series. The results of training of multilayer perceptrons showed that perceptrons with architectures of 3-2-1 within the Freeze-up phase (arithmetic mean of the mean square errors for training epoch $\overline {MSE} = 0.0155$ ) and 3-6-1 within the Break-up phase ( $\overline {MSE} = 0.0105$ ) have the least mean-squared error for the last training epoch. Tests within the holdout samples prove that multilayer perceptrons give more adequate and reliable prediction of the ice coverage of Lake Onega (mean-squared prediction error MSPE = 0.0076) comparing with statistical methods such as linear regression, moving average and autoregressive analyses of the first and second order.

APA, Harvard, Vancouver, ISO, and other styles

22

Erdogmus, D., O. Fontenla-Romero, J. C. Principe, A. Alonso-Betanzos, and E. Castillo. "Linear-Least-Squares Initialization of Multilayer Perceptrons Through Backpropagation of the Desired Response." IEEE Transactions on Neural Networks 16, no. 2 (March 2005): 325–37. http://dx.doi.org/10.1109/tnn.2004.841777.

Full text

APA, Harvard, Vancouver, ISO, and other styles

23

Back, Andrew D., and Ah Chung Tsoi. "An Adaptive Lattice Architecture for Dynamic Multilayer Perceptrons." Neural Computation 4, no. 6 (November 1992): 922–31. http://dx.doi.org/10.1162/neco.1992.4.6.922.

Full text

Abstract:

Time-series modeling is a topic of growing interest in neural network research. Various methods have been proposed for extending the nonlinear approximation capabilities to time-series modeling problems. A multilayer perceptron (MLP) with a global-feedforward local-recurrent structure was recently introduced as a new approach to modeling dynamic systems. The network uses adaptive infinite impulse response (IIR) synapses (it is thus termed an IIR MLP), and was shown to have good modeling performance. One problem with linear IIR filters is that the rate of convergence depends on the covariance matrix of the input data. This extends to the IIR MLP: it learns well for white input signals, but converges more slowly with nonwhite inputs. To solve this problem, the adaptive lattice multilayer perceptron (AL MLP), is introduced. The network structure performs Gram-Schmidt orthogonalization on the input data to each synapse. The method is based on the same principles as the Gram-Schmidt neural net proposed by Orfanidis (1990b), but instead of using a network layer for the orthogonalization, each synapse comprises an adaptive lattice filter. A learning algorithm is derived for the network that minimizes a mean square error criterion. Simulations are presented to show that the network architecture significantly improves the learning rate when correlated input signals are present.

APA, Harvard, Vancouver, ISO, and other styles

24

Back, Andrew D., and Ah Chung Tsoi. "A Simplified Gradient Algorithm for IIR Synapse Multilayer Perceptrons." Neural Computation 5, no. 3 (May 1993): 456–62. http://dx.doi.org/10.1162/neco.1993.5.3.456.

Full text

Abstract:

A network architecture with a global feedforward local recurrent construction was presented recently as a new means of modeling nonlinear dynamic time series (Back and Tsoi 1991a). The training rule used was based on minimizing the least mean square (LMS) error and performed well, although the amount of memory required for large networks may become significant if a large number of feedback connections are used. In this note, a modified training algorithm based on a technique for linear filters is presented, simplifying the gradient calculations significantly. The memory requirements are reduced from O[na(na + nb)Ns] to O[(2na + nb)Ns], where na is the number of feedback delays, and Ns is the total number of synapses. The new algorithm reduces the number of multiply-adds needed to train each synapse by na at each time step. Simulations indicate that the algorithm has almost identical performance to the previous one.

APA, Harvard, Vancouver, ISO, and other styles

25

NABNEY, IAN T. "EFFICIENT TRAINING OF RBF NETWORKS FOR CLASSIFICATION." International Journal of Neural Systems 14, no. 03 (June 2004): 201–8. http://dx.doi.org/10.1142/s0129065704001930.

Full text

Abstract:

Radial Basis Function networks with linear outputs are often used in regression problems because they can be substantially faster to train than Multi-layer Perceptrons. For classification problems, the use of linear outputs is less appropriate as the outputs are not guaranteed to represent probabilities. We show how RBFs with logistic and softmax outputs can be trained efficiently using the Fisher scoring algorithm. This approach can be used with any model which consists of a generalised linear output function applied to a model which is linear in its parameters. We compare this approach with standard non-linear optimisation algorithms on a number of datasets.

APA, Harvard, Vancouver, ISO, and other styles

26

Amari, Shun-ichi. "Natural Gradient Works Efficiently in Learning." Neural Computation 10, no. 2 (February 1, 1998): 251–76. http://dx.doi.org/10.1162/089976698300017746.

Full text

Abstract:

When a parameter space has a certain underlying structure, the ordinary gradient of a function does not represent its steepest direction, but the natural gradient does. Information geometry is used for calculating the natural gradients in the parameter space of perceptrons, the space of matrices (for blind source separation), and the space of linear dynamical systems (for blind source deconvolution). The dynamical behavior of natural gradient online learning is analyzed and is proved to be Fisher efficient, implying that it has asymptotically the same performance as the optimal batch estimation of parameters. This suggests that the plateau phenomenon, which appears in the backpropagation learning algorithm of multilayer perceptrons, might disappear or might not be so serious when the natural gradient is used. An adaptive method of updating the learning rate is proposed and analyzed.

APA, Harvard, Vancouver, ISO, and other styles

27

Cazé, Romain D., and Marcel Stimberg. "Neurons with dendrites can perform linearly separable computations with low resolution synaptic weights." F1000Research 9 (April 18, 2021): 1174. http://dx.doi.org/10.12688/f1000research.26486.3.

Full text

Abstract:

In theory, neurons modelled as single layer perceptrons can implement all linearly separable computations. In practice, however, these computations may require arbitrarily precise synaptic weights. This is a strong constraint since both biological neurons and their artificial counterparts have to cope with limited precision. Here, we explore how non-linear processing in dendrites helps overcome this constraint. We start by finding a class of computations which requires increasing precision with the number of inputs in a perceptron and show that it can be implemented without this constraint in a neuron with sub-linear dendritic subunits. Then, we complement this analytical study by a simulation of a biophysical neuron model with two passive dendrites and a soma, and show that it can implement this computation. This work demonstrates a new role of dendrites in neural computation: by distributing the computation across independent subunits, the same computation can be performed more efficiently with less precise tuning of the synaptic weights. This work not only offers new insight into the importance of dendrites for biological neurons, but also paves the way for new, more efficient architectures of artificial neuromorphic chips.

APA, Harvard, Vancouver, ISO, and other styles

28

Cazé, Romain D., and Marcel Stimberg. "Neurons with dendrites can perform linearly separable computations with low resolution synaptic weights." F1000Research 9 (January 20, 2021): 1174. http://dx.doi.org/10.12688/f1000research.26486.2.

Full text

Abstract:

In theory, neurons modelled as single layer perceptrons can implement all linearly separable computations. In practice, however, these computations may require arbitrarily precise synaptic weights. This is a strong constraint since both biological neurons and their artificial counterparts have to cope with limited precision. Here, we explore how non-linear processing in dendrites helps overcome this constraint. We start by finding a class of computations which requires increasing precision with the number of inputs in a perceptron and show that it can be implemented without this constraint in a neuron with sub-linear dendritic subunits. Then, we complement this analytical study by a simulation of a biophysical neuron model with two passive dendrites and a soma, and show that it can implement this computation. This work demonstrates a new role of dendrites in neural computation: by distributing the computation across independent subunits, the same computation can be performed more efficiently with less precise tuning of the synaptic weights. This work not only offers new insight into the importance of dendrites for biological neurons, but also paves the way for new, more efficient architectures of artificial neuromorphic chips.

APA, Harvard, Vancouver, ISO, and other styles

29

Cazé, Romain D., and Marcel Stimberg. "Dendritic neurons can perform linearly separable computations with low resolution synaptic weights." F1000Research 9 (September 28, 2020): 1174. http://dx.doi.org/10.12688/f1000research.26486.1.

Full text

Abstract:

In theory, neurons modelled as single layer perceptrons can implement all linearly separable computations. In practice, however, these computations may require arbitrarily precise synaptic weights. This is a strong constraint since both biological neurons and their artificial counterparts have to cope with limited precision. Here, we explore how non-linear processing in dendrites helps overcome this constraint. We start by finding a class of computations which requires increasing precision with the number of inputs in a Perceptron and show that it can be implemented without this constraint in a neuron with sub-linear dendritic subunits. Then, we complement this analytical study by a simulation of a biophysical neuron model with two passive dendrites and a soma, and show that it can implement this computation. This work demonstrates a new role of dendrites in neural computation: by distributing the computation across independent subunits, the same computation can be performed more efficiently with less precise tuning of the synaptic weights. This work not only offers new insight into the importance of dendrites for biological neurons, but also paves the way for new, more efficient architectures of artificial neuromorphic chips.

APA, Harvard, Vancouver, ISO, and other styles

30

Hayasaka, Taichi, Masashi Kitahara, and Shiro Usui. "On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models." Neural Computation 16, no. 1 (January 1, 2004): 99–114. http://dx.doi.org/10.1162/08997660460734010.

Full text

Abstract:

In order to analyze the stochastic property of multilayered perceptrons or other learning machines, we deal with simpler models and derive the asymptotic distribution of the least-squares estimators of their parameters. In the case where a model is unidentified, we show different results from traditional linear models: the well-known property of asymptotic normality never holds for the estimates of redundant parameters.

APA, Harvard, Vancouver, ISO, and other styles

31

John Ohabuiro, Ugochukwu O. Matthew, Salisu Umar, Danladi Agadi Tonga, and Amaonwu Onyebuchi. "Global Solar Radiation Modelling using an Artificial Neural Network for Kazaure, Jigawa State, Nigeria." December 2022 4, no. 4 (January 21, 2023): 316–31. http://dx.doi.org/10.36548/jeea.2022.4.008.

Full text

Abstract:

This research presents an algorithm based on Artificial Neural Networks (ANN), for estimating monthly mean daily and hourly values of solar global radiation. To effectively investigate solar energy consumption and estimate solar renewable energy resources, the Hourly Global Solar Radiation measurements are necessary. In order to predict monthly average daily global sun irradiance on a horizontal area of Kazaure- Nigeria, this study creates a model utilizing ANN to solve the problem of solar energy distribution. Five empirical correlations are developed using the data from 42 months to aid in the prediction of the solar energy distribution pattern. The software is constructed around the Multilayer Perceptron under categorized tabs, with Multilayer perception in neural network Toolbox in MATLAB 9.7 version as a feed forward ANN that maps sets of input data into a set of suitable output. It differs from conventional linear perception by employing three or more layers of neurons (nodes) with nonlinear activation functions. It is also more effective than perceptrons in identifying input that is not linearly separable by a linear hyper-plane. Results obtained utilizing the suggested structure reveals good agreement between the calculated and measured levels of global solar irradiation. The ANN model is shown to be superior when compared to empirical models, due to negligible noise margin.

APA, Harvard, Vancouver, ISO, and other styles

32

Wu, Buchen, and Jiwei Qin. "A List-Ranking Framework Based on Linear and Non-Linear Fusion for Recommendation from Implicit Feedback." Entropy 24, no. 6 (May 31, 2022): 778. http://dx.doi.org/10.3390/e24060778.

Full text

Abstract:

Although most list-ranking frameworks are based on multilayer perceptrons (MLP), they still face limitations within the method itself in the field of recommender systems in two respects: (1) MLP suffer from overfitting when dealing with sparse vectors. At the same time, the model itself tends to learn in-depth features of user–item interaction behavior but ignores some low-rank and shallow information present in the matrix. (2) Existing ranking methods cannot effectively deal with the problem of ranking between items with the same rating value and the problem of inconsistent independence in reality. We propose a list ranking framework based on linear and non-linear fusion for recommendation from implicit feedback, named RBLF. First, the model uses dense vectors to represent users and items through one-hot encoding and embedding. Second, to jointly learn shallow and deep user–item interaction, we use the interaction grabbing layer to capture the user–item interaction behavior through dense vectors of users and items. Finally, RBLF uses the Bayesian collaborative ranking to better fit the characteristics of implicit feedback. Eventually, the experiments show that the performance of RBLF obtains a significant improvement.

APA, Harvard, Vancouver, ISO, and other styles

33

Khoshaim, Ahmed B., Essam B. Moustafa, Omar Talal Bafakeeh, and Ammar H. Elsheikh. "An Optimized Multilayer Perceptrons Model Using Grey Wolf Optimizer to Predict Mechanical and Microstructural Properties of Friction Stir Processed Aluminum Alloy Reinforced by Nanoparticles." Coatings 11, no. 12 (November 30, 2021): 1476. http://dx.doi.org/10.3390/coatings11121476.

Full text

Abstract:

In the current investigation, AA2024 aluminum alloy is reinforced by alumina nanoparticles using a friction stir process (FSP) with multiple passes. The mechanical properties and microstructure observation are conducted experimentally using tensile, microhardness, and microscopy analysis methods. The impacts of the process parameters on the output responses, such as mechanical properties and microstructure grain refinement, were investigated. The effect of multiple FSP passes on the grain refinement, and various mechanical properties are evaluated, then the results are conducted to train a hybrid artificial intelligence predictive model. The model consists of a multilayer perceptrons optimized by a grey wolf optimizer to predict mechanical and microstructural properties of friction stir processed aluminum alloy reinforced by alumina nanoparticles. The inputs of the model were rotational speed, linear processing speed, and number of passes; while the outputs were grain size, aspect ratio, microhardness, and ultimate tensile strength. The prediction accuracy of the developed hybrid model was compared with that of standalone multilayer perceptrons model using different error measures. The developed hybrid model shows a higher accuracy compared with the standalone model.

APA, Harvard, Vancouver, ISO, and other styles

34

Roj, J. "Neural Network Based Real-time Correction of Transducer Dynamic Errors." Measurement Science Review 13, no. 6 (December 1, 2013): 286–91. http://dx.doi.org/10.2478/msr-2013-0042.

Full text

Abstract:

Abstract In order to carry out real-time dynamic error correction of transducers described by a linear differential equation, a novel recurrent neural network was developed. The network structure is based on solving this equation with respect to the input quantity when using the state variables. It is shown that such a real-time correction can be carried out using simple linear perceptrons. Due to the use of a neural technique, knowledge of the dynamic parameters of the transducer is not necessary. Theoretical considerations are illustrated by the results of simulation studies performed for the modeled second order transducer. The most important properties of the neural dynamic error correction, when emphasizing the fundamental advantages and disadvantages, are discussed.

APA, Harvard, Vancouver, ISO, and other styles

35

Luna, Antonio Madueño, Miriam López Lineros, Javier Estévez Gualda, Juan Vicente Giráldez Cervera, and José Miguel Madueño Luna. "Assessing the Best Gap-Filling Technique for River Stage Data Suitable for Low Capacity Processors and Real-Time Application Using IoT." Sensors 20, no. 21 (November 7, 2020): 6354. http://dx.doi.org/10.3390/s20216354.

Full text

Abstract:

Hydrometeorological data sets are usually incomplete due to different reasons (malfunctioning sensors, collected data storage problems, etc.). Missing data do not only affect the resulting decision-making process, but also the choice of a particular analysis method. Given the increase of extreme events due to climate change, it is necessary to improve the management of water resources. Due to the solution of this problem requires the development of accurate estimations and its application in real time, this work present two contributions. Firstly, different gap-filling techniques have been evaluated in order to select the most adequate one for river stage series: (i) cubic splines (CS), (ii) radial basis function (RBF) and (iii) multilayer perceptron (MLP) suitable for small processors like Arduino or Raspberry Pi. The results obtained confirmed that splines and monolayer perceptrons had the best performances. Secondly, a pre-validating Internet of Things (IoT) device was developed using a dynamic seed non-linear autoregressive neural network (NARNN). This automatic pre-validation in real time was tested satisfactorily, sending the data to the catchment basin process center (CPC) by using remote communication based on 4G technology.

APA, Harvard, Vancouver, ISO, and other styles

36

Legenstein, Robert, Christian Naeger, and Wolfgang Maass. "What Can a Neuron Learn with Spike-Timing-Dependent Plasticity?" Neural Computation 17, no. 11 (November 1, 2005): 2337–82. http://dx.doi.org/10.1162/0899766054796888.

Full text

Abstract:

Spiking neurons are very flexible computational modules, which can implement with different values of their adjustable synaptic parameters an enormous variety of different transformations F from input spike trains to output spike trains. We examine in this letter the question to what extent a spiking neuron with biologically realistic models for dynamic synapses can be taught via spike-timing-dependent plasticity (STDP) to implement a given transformation F. We consider a supervised learning paradigm where during training, the output of the neuron is clamped to the target signal (teacher forcing). The well-known perceptron convergence theorem asserts the convergence of a simple supervised learning algorithm for drastically simplified neuron models (McCulloch-Pitts neurons). We show that in contrast to the perceptron convergence theorem, no theoretical guarantee can be given for the convergence of STDP with teacher forcing that holds for arbitrary input spike patterns. On the other hand, we prove that average case versions of the perceptron convergence theorem hold for STDP in the case of uncorrelated and correlated Poisson input spike trains and simple models for spiking neurons. For a wide class of cross-correlation functions of the input spike trains, the resulting necessary and sufficient condition can be formulated in terms of linear separability, analogously as the well-known condition of learnability by perceptrons. However, the linear separability criterion has to be applied here to the columns of the correlation matrix of the Poisson input. We demonstrate through extensive computer simulations that the theoretically predicted convergence of STDP with teacher forcing also holds for more realistic models for neurons, dynamic synapses, and more general input distributions. In addition, we show through computer simulations that these positive learning results hold not only for the common interpretation of STDP, where STDP changes the weights of synapses, but also for a more realistic interpretation suggested by experimental data where STDP modulates the initial release probability of dynamic synapses.

APA, Harvard, Vancouver, ISO, and other styles

37

Kerlirzin, P., and P. Refregier. "Theoretical investigation of the robustness of multilayer perceptrons: analysis of the linear case and extension to nonlinear networks." IEEE Transactions on Neural Networks 6, no. 3 (May 1995): 560–71. http://dx.doi.org/10.1109/72.377963.

Full text

APA, Harvard, Vancouver, ISO, and other styles

38

Tsaptsinos, D., and J. R. Leigh. "Modelling of a fermentation process using Multi-Layer Perceptrons: Epochs vs Pattern learning, Sigmoid vs Linear transfer function." Journal of Microcomputer Applications 16, no. 2 (April 1993): 125–36. http://dx.doi.org/10.1006/jmca.1993.1011.

Full text

APA, Harvard, Vancouver, ISO, and other styles

39

Barnawi, Abdulaziz Y., and Ismail M. Keshta. "Energy Management in Wireless Sensor Networks Based on Naive Bayes, MLP, and SVM Classifications: A Comparative Study." Journal of Sensors 2016 (2016): 1–12. http://dx.doi.org/10.1155/2016/6250319.

Full text

Abstract:

Maximizing wireless sensor networks (WSNs) lifetime is a primary objective in the design of these networks. Intelligent energy management models can assist designers to achieve this objective. These models aim to reduce the number of selected sensors to report environmental measurements and, hence, achieve higher energy efficiency while maintaining the desired level of accuracy in the reported measurement. In this paper, we present a comparative study of three intelligent models based on Naive Bayes, Multilayer Perceptrons (MLP), and Support Vector Machine (SVM) classifiers. Simulation results show that Linear-SVM selects sensors that produce higher energy efficiency compared to those selected by MLP and Naive Bayes for the same WSNs Lifetime Extension Factor.

APA, Harvard, Vancouver, ISO, and other styles

40

Coutinho, Eluã Ramos, Robson Mariano Silva, and Angel Ramon Sanchez Delgado. "Utilização de Técnicas de Inteligência Computacional na Predição de Dados Meteorológicos." Revista Brasileira de Meteorologia 31, no. 1 (March 2016): 24–36. http://dx.doi.org/10.1590/0102-778620140115.

Full text

Abstract:

Resumo O presente trabalho apresenta uma abordagem computacional para a predição de um passo à frente em séries de dados meteorológicos pertencentes às regiões de Paty do Alferes e Paracambi, situadas no estado do Rio de Janeiro (RJ). Para tanto, foram utilizados dois modelos de Redes Neurais Artificiais (RNAs): Perceptrons de Múltiplas Camadas (MLP) e Função de Base Radial (RBF). Para confirmar o desempenho dos modelos foi realizada a predição de variáveis horárias e mensais, que foram comparadas com resultados obtidos por modelos de Regressão Linear Múltipla (RLM), confrontadas com os dados registrados pelas estações meteorológicas e analisadas por meio de técnicas estatísticas, apresentando resultados favoráveis entre 91% a 96% de acerto para todos os casos. Além disso, as previsões também demonstraram uma forte correlação linear com os dados registrados, mantendo-se entre 0,61 a 0,94. Como resultado, pode se destacar as RNAs como uma forte ferramenta para predição dos dados meteorológicos analisados.

APA, Harvard, Vancouver, ISO, and other styles

41

Silva, Cecilia Cordeiro da, Clarisse Lins de Lima, Ana Clara Gomes da Silva, Giselle Machado Magalhães Moreno, Anwar Musah, Aisha Aldosery, Livia Dutra, et al. "Forecasting Dengue, Chikungunya and Zika cases in Recife, Brazil: a spatio-temporal approach based on climate conditions, health notifications and machine learning." Research, Society and Development 10, no. 12 (September 26, 2021): e452101220804. http://dx.doi.org/10.33448/rsd-v10i12.20804.

Full text

Abstract:

Dengue has become a challenge for many countries. Arboviruses transmitted by Aedes aegypti spread rapidly over the last decades. The emergence chikungunya fever and zika in South America poses new challenges to vector monitoring and control. This situation got worse from 2015 and 2016, with the rapid spread of chikungunya, causing fever and muscle weakness, and Zika virus, related to cases of microcephaly in newborns and the occurrence of Guillain-Barret syndrome, an autoimmune disease that affects the nervous system. The objective of this work was to construct a tool to forecast the distribution of arboviruses transmitted by the mosquito Aedes aegypti by implementing dengue, zika and chikungunya transmission predictors based on machine learning, focused on multilayer perceptrons neural networks, support vector machines and linear regression models. As a case study, we investigated forecasting models to predict the spatio-temporal distribution of cases from primary health notification data and climate variables (wind velocity, temperature and pluviometry) from Recife, Brazil, from 2013 to 2016, including 2015’s outbreak. The use of spatio-temporal analysis over multilayer perceptrons and support vector machines results proved to be very effective in predicting the distribution of arbovirus cases. The models indicate that the southern and western regions of Recife were very susceptible to outbreaks in the period under investigation. The proposed approach could be useful to support health managers and epidemiologists to prevent outbreaks of arboviruses transmitted by Aedes aegypti and promote public policies for health promotion and sanitation.

APA, Harvard, Vancouver, ISO, and other styles

42

SAKURAI, AKITO. "A FAST AND CONVERGENT STOCHASTIC MLP LEARNING ALGORITHM." International Journal of Neural Systems 11, no. 06 (December 2001): 573–83. http://dx.doi.org/10.1142/s0129065701000977.

Full text

Abstract:

We propose a stochastic learning algorithm for multilayer perceptrons of linear-threshold function units, which theoretically converges with probability one and experimentally exhibits 100% convergence rate and remarkable speed on parity and classification problems with typical generalization accuracy. For learning the n bit parity function with n hidden units, the algorithm converged on all the trials we tested (n=2 to 12) after 5.8· 4.1n presentations for 0.23· 4.0n-6 seconds on a 533MHz Alpha 21164A chip on average, which is five to ten times faster than Levenberg-Marquardt algorithm with restarts. For a medium size classification problem known as Thyroid in UCI repository, the algorithm is faster in speed and comparative in generalization accuracy than the standard backpropagation and Levenberg-Marquardt algorithms.

APA, Harvard, Vancouver, ISO, and other styles

43

XU, LEI, STAN KLASA, and ALAN YUILLE. "RECENT ADVANCES ON TECHNIQUES OF STATIC FEEDFORWARD NETWORKS WITH SUPERVISED LEARNING." International Journal of Neural Systems 03, no. 03 (January 1992): 253–90. http://dx.doi.org/10.1142/s0129065792000218.

Full text

Abstract:

The rediscovery and popularization of the backpropagation training technique for multilayer perceptrons as well as the invention of the Boltzmann machine learning algorithm has given a new boost to the study on supervised learning networks. In recent years, besides widely spread applications and various further improvements of the classical backpropagation technique, many new supervised learning models, techniques as well as theories, have also been proposed in a vast number of publications. This paper tries to give a rather systematic review on the recent advances on supervised learning techniques and models for static feedforward networks. We summarize a great number of developments into four aspects: (1) Various improvements and variants made on the classical backpropagation techniques for Multilayer(static) perceptron nets, for speeding up training, avoiding local minima, increasing the generalization ability as well as for many other interesting purposes. (2) A number of other learning methods for training multilayer (static) perceptron, such as derivative estimation by perturbation, direct weight update by perturbation, genetic algorithms, recursive least square estimate and extended Kalman filters, linear programming, the policy of fixing one layer while updating another, constructing networks by converting decision tree classifiers and others. (3) Various other feedforward models which are also able to implement function approximation, probability density estimation and classification, including various models of basis function expansion (e.g. radial basis functions, restricted coulomb energy, multivariate adaptive regression splines, trigonometric and polynomial bases, projection pursuit, basis function tree and many others) and several other supervised learning models. (4) Models with complex structures, e.g. modular architecture, hierarchy architecture and others. Altogether, we try to give a global picture of the present state of supervised learning techniques (not including all the theoretical developments) for training static feedforward networks.

APA, Harvard, Vancouver, ISO, and other styles

44

Karras, D. A., and B. G. Mertzios. "Time series modeling of endocardial border motion in ultrasonic images comparing support vector machines, multilayer perceptrons and linear estimation techniques." Measurement 36, no. 3-4 (October 2004): 331–45. http://dx.doi.org/10.1016/j.measurement.2004.09.012.

Full text

APA, Harvard, Vancouver, ISO, and other styles

45

Farrell, Max H., Tengyuan Liang, and Sanjog Misra. "Deep Neural Networks for Estimation and Inference." Econometrica 89, no. 1 (2021): 181–213. http://dx.doi.org/10.3982/ecta16901.

Full text

Abstract:

We study deep neural networks and their use in semiparametric inference. We establish novel nonasymptotic high probability bounds for deep feedforward neural nets. These deliver rates of convergence that are sufficiently fast (in some cases minimax optimal) to allow us to establish valid second‐step inference after first‐step estimation with deep learning, a result also new to the literature. Our nonasymptotic high probability bounds, and the subsequent semiparametric inference, treat the current standard architecture: fully connected feedforward neural networks (multilayer perceptrons), with the now‐common rectified linear unit activation function, unbounded weights, and a depth explicitly diverging with the sample size. We discuss other architectures as well, including fixed‐width, very deep networks. We establish the nonasymptotic bounds for these deep nets for a general class of nonparametric regression‐type loss functions, which includes as special cases least squares, logistic regression, and other generalized linear models. We then apply our theory to develop semiparametric inference, focusing on causal parameters for concreteness, and demonstrate the effectiveness of deep learning with an empirical application to direct mail marketing.

APA, Harvard, Vancouver, ISO, and other styles

46

Tsuchida, Russell, Tim Pearce, Chris Van der Heide, Fred Roosta, and Marcus Gallagher. "Avoiding Kernel Fixed Points: Computing with ELU and GELU Infinite Networks." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 11 (May 18, 2021): 9967–77. http://dx.doi.org/10.1609/aaai.v35i11.17197.

Full text

Abstract:

Analysing and computing with Gaussian processes arising from infinitely wide neural networks has recently seen a resurgence in popularity. Despite this, many explicit covariance functions of networks with activation functions used in modern networks remain unknown. Furthermore, while the kernels of deep networks can be computed iteratively, theoretical understanding of deep kernels is lacking, particularly with respect to fixed-point dynamics. Firstly, we derive the covariance functions of multi-layer perceptrons (MLPs) with exponential linear units (ELU) and Gaussian error linear units (GELU) and evaluate the performance of the limiting Gaussian processes on some benchmarks. Secondly, and more generally, we analyse the fixed-point dynamics of iterated kernels corresponding to a broad range of activation functions. We find that unlike some previously studied neural network kernels, these new kernels exhibit non-trivial fixed-point dynamics which are mirrored in finite-width neural networks. The fixed point behaviour present in some networks explains a mechanism for implicit regularisation in overparameterised deep models. Our results relate to both the static iid parameter conjugate kernel and the dynamic neural tangent kernel constructions

APA, Harvard, Vancouver, ISO, and other styles

47

SRIRAAM, N., and C. ESWARAN. "EEG SIGNAL COMPRESSION USING RADIAL BASIS NEURAL NETWORKS." Journal of Mechanics in Medicine and Biology 04, no. 02 (June 2004): 143–52. http://dx.doi.org/10.1142/s0219519404000928.

Full text

Abstract:

This paper describes a two-stage lossless compression scheme for electroencephalographic (EEG) signals using radial basis neural network predictors. Two variants of the radial basis network, namely, the radial basis function network and the generalized regression neural network are used in the first stage and their performances are evaluated in terms of compression ratio. The training is imparted to the network by using two training schemes, namely, single block scheme and block adaptive scheme. The compression ratios achieved by these networks when used along with arithmetic encoders in a two-stage compression scheme are obtained for different EEG test files. It is found that the generalized regression neural network performs better than other neural network models such as multilayer perceptrons and Elman network and linear predictor such as FIR.

APA, Harvard, Vancouver, ISO, and other styles

48

KUTSCHENREITER-PRASZKIEWICZ, Izabela. "MACHINE LEARNING IN SMED." Journal of Machine Engineering 18, no. 2 (June 12, 2018): 31–40. http://dx.doi.org/10.5604/01.3001.0012.0923.

Full text

Abstract:

The paper discusses Single Minute Exchange of Die (SMED) and machine learning methods, such as neural networks and a decision tree. SMED is one of lean production methods for reducing waste in the manufacturing process, which helps to reorganize a conversion of the manufacturing process from current to the next product. SMED needs set-up activity analyses, which include activity classification, working time measurement and work improvement. The analyses presented in the article are focused on selecting the time measurement method useful from the SMED perspective. Time measurement methods and their comparison are presented in the paper. Machine learning methods are used to suggest the method of time measurement which should be applied in a particular case of workstation reorganization. A training set is developed and an example of classification is presented. Time and motion study is one of important methods of estimating machine changeover time. In the field of time study, researchers present the obtained results by using (linear) multi-linear regression models (MLR), and (non-linear) multi-layer perceptrons (MLP). The presented approach is particularly important for the enterprises which offer make-to-order products. Development of the SMED method can influence manufacturing cost reduction of customized products. In variety oriented manufacturing, SMED supports flexibility and adaptability of the manufacturing system.

APA, Harvard, Vancouver, ISO, and other styles

49

GHOSH, JOYDEEP, and YOAN SHIN. "EFFICIENT HIGHER-ORDER NEURAL NETWORKS FOR CLASSIFICATION AND FUNCTION APPROXIMATION." International Journal of Neural Systems 03, no. 04 (January 1992): 323–50. http://dx.doi.org/10.1142/s0129065792000255.

Full text

Abstract:

This paper introduces a class of higher-order networks called pi-sigma networks (PSNs). PSNs are feedforward networks with a single “hidden” layer of linear summing units and with product units in the output layer. A PSN uses these product units to indirectly incorporate the capabilities of higher-order networks while greatly reducing network complexity. PSNs have only one layer of adjustable weights and exhibit fast learning. A PSN with K summing units provides a constrained Kth order approximation of a continuous function. A generalization of the PSN is presented that can uniformly approximate any continuous function defined on a compact set. The use of linear hidden units makes it possible to mathematically study the convergence properties of various LMS type learning algorithms for PSNs. We show that it is desirable to update only a partial set of weights at a time rather than synchronously updating all the weights. Bounds for learning rates which guarantee convergence are derived. Several simulation results on pattern classification and function approximation problems highlight the capabilities of the PSN. Extensive comparisons are made with other higher order networks and with multilayered perceptrons. The neurobiological plausibility of PSN type networks is also discussed.

APA, Harvard, Vancouver, ISO, and other styles

50

SUYKENS, JOHAN A. K., JOOS VANDEWALLE, and BART DE MOOR. "INTELLIGENCE AND COOPERATIVE SEARCH BY COUPLED LOCAL MINIMIZERS." International Journal of Bifurcation and Chaos 11, no. 08 (August 2001): 2133–44. http://dx.doi.org/10.1142/s0218127401003371.

Full text

Abstract:

We show how coupling of local optimization processes can lead to better solutions than multistart local optimization consisting of independent runs. This is achieved by minimizing the average energy cost of the ensemble, subject to synchronization constraints between the state vectors of the individual local minimizers. From an augmented Lagrangian which incorporates the synchronization constraints both as soft and hard constraints, a network is derived wherein the local minimizers interact and exchange information through the synchronization constraints. From the viewpoint of neural networks, the array can be considered as a Lagrange programming network for continuous optimization and as a cellular neural network (CNN). The penalty weights associated with the soft state synchronization constraints follow from the solution to a linear program. This shows that the energy cost of the ensemble should maximally decrease. In this way successful local minimizers can implicitly impose their state to the others through a mechanism of master–slave dynamics resulting in a cooperative search mechanism. Improved information spreading within the ensemble is obtained by applying the concept of small-world networks. We illustrate the new optimization method on two different problems: supervized learning of multilayer perceptrons and optimization of Lennard–Jones clusters. The initial distribution of the local minimizers plays an important role. For the training of multilayer perceptrons this is related to the choice of the prior on the interconnection weights in Bayesian learning methods. Depending on the choice of this initial distribution, coupled local minimizers (CLM) can avoid overfitting and produce good generalization, i.e. reach a state of intelligence. In potential energy surface optimization of Lennard–Jones clusters, this choice is equally important. In this case it can be related to considering a confining potential. This work suggests, in an interdisciplinary context, the importance of information exchange and state synchronization within ensembles, towards issues such as evolution, collective behavior, optimality and intelligence.

APA, Harvard, Vancouver, ISO, and other styles

Journal articles on the topic 'Linear perceptrons'

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles