Journal articles on the topic 'Neural networks (NNs)'
To see the other types of publications on this topic, follow the link: Neural networks (NNs).
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 'Neural networks (NNs).'
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
Thakur, Amey. "Fundamentals of Neural Networks." International Journal for Research in Applied Science and Engineering Technology 9, no. VIII (August 15, 2021): 407–26. http://dx.doi.org/10.22214/ijraset.2021.37362.
Full textAbstract:
The purpose of this study is to familiarise the reader with the foundations of neural networks. Artificial Neural Networks (ANNs) are algorithm-based systems that are modelled after Biological Neural Networks (BNNs). Neural networks are an effort to use the human brain's information processing skills to address challenging real-world AI issues. The evolution of neural networks and their significance are briefly explored. ANNs and BNNs are contrasted, and their qualities, benefits, and disadvantages are discussed. The drawbacks of the perceptron model and their improvement by the sigmoid neuron and ReLU neuron are briefly discussed. In addition, we give a bird's-eye view of the different Neural Network models. We study neural networks (NNs) and highlight the different learning approaches and algorithms used in Machine Learning and Deep Learning. We also discuss different types of NNs and their applications. A brief introduction to Neuro-Fuzzy and its applications with a comprehensive review of NN technological advances is provided.
2
Jwo, Dah-Jing, and Chien-Cheng Lai. "Neural Network-Based Geometry Classification for Navigation Satellite Selection." Journal of Navigation 56, no. 2 (May 2003): 291–304. http://dx.doi.org/10.1017/s0373463303002200.
Full textAbstract:
The neural networks (NN)-based geometry classification for good or acceptable navigation satellite subset selection is presented. The approach is based on classifying the values of satellite Geometry Dilution of Precision (GDOP) utilizing the classification-type NNs. Unlike some of the NNs that approximate the function, such as the back-propagation neural network (BPNN), the NNs here are employed as classifiers. Although BPNN can also be employed as a classifier, it takes a long training time. Two other methods that feature a fast learning speed will be implemented, including Optimal Interpolative (OI) Net and Probabilistic Neural Network (PNN). Simulation results from these three neural networks are presented. The classification performance and computational expense of neural network-based GDOP classification are explored.
3
Guidotti, Dario. "Verification and Repair of Neural Networks." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 18 (May 18, 2021): 15714–15. http://dx.doi.org/10.1609/aaai.v35i18.17854.
Full textAbstract:
Neural Networks (NNs) are popular machine learning models which have found successful application in many different domains across computer science. However, it is hard to provide any formal guarantee on the behaviour of neural networks and therefore their reliability is still in doubt, especially concerning their deployment in safety and security-critical applications. Verification emerged as a promising solution to address some of these problems. In the following, I will present some of my recent efforts in verifying NNs.
4
IKEDA, TAKASHI, and MASAFUMI HAGIWARA. "CONTENT-BASED IMAGE RETRIEVAL SYSTEM USING NEURAL NETWORKS." International Journal of Neural Systems 10, no. 05 (October 2000): 417–24. http://dx.doi.org/10.1142/s0129065700000326.
Full textAbstract:
An effective image retrieval system is developed based on the use of neural networks (NNs). It takes advantages of association ability of multilayer NNs as matching engines which calculate similarities between a user's drawn sketch and the stored images. The NNs memorize pixel information of every size-reduced image (thumbnail) in the learning phase. In the retrieval phase, pixel information of a user's drawn rough sketch is inputted to the learned NNs and they estimate the candidates. Thus the system can retrieve candidates quickly and correctly by utilizing the parallelism and association ability of NNs. In addition, the system has learning capability: it can automatically extract features of a user's drawn sketch during the retrieval phase and can store them as additional information to improve the performance. The software for querying, including efficient graphical user interfaces, has been implemented and tested. The effectiveness of the proposed system has been investigated through various experimental tests.
5
Ghorbani, Behrooz, Song Mei, Theodor Misiakiewicz, and Andrea Montanari. "When do neural networks outperform kernel methods?*." Journal of Statistical Mechanics: Theory and Experiment 2021, no. 12 (December 1, 2021): 124009. http://dx.doi.org/10.1088/1742-5468/ac3a81.
Full textAbstract:
Abstract For a certain scaling of the initialization of stochastic gradient descent (SGD), wide neural networks (NN) have been shown to be well approximated by reproducing kernel Hilbert space (RKHS) methods. Recent empirical work showed that, for some classification tasks, RKHS methods can replace NNs without a large loss in performance. On the other hand, two-layers NNs are known to encode richer smoothness classes than RKHS and we know of special examples for which SGD-trained NN provably outperform RKHS. This is true even in the wide network limit, for a different scaling of the initialization. How can we reconcile the above claims? For which tasks do NNs outperform RKHS? If covariates are nearly isotropic, RKHS methods suffer from the curse of dimensionality, while NNs can overcome it by learning the best low-dimensional representation. Here we show that this curse of dimensionality becomes milder if the covariates display the same low-dimensional structure as the target function, and we precisely characterize this tradeoff. Building on these results, we present the spiked covariates model that can capture in a unified framework both behaviors observed in earlier work. We hypothesize that such a latent low-dimensional structure is present in image classification. We test numerically this hypothesis by showing that specific perturbations of the training distribution degrade the performances of RKHS methods much more significantly than NNs.
6
Đerek, Jurica, Marjan Sikora, Luka Kraljević, and Mladen Russo. "Using Neural Networks for Bicycle Route Planning." Applied Sciences 11, no. 21 (October 27, 2021): 10065. http://dx.doi.org/10.3390/app112110065.
Full textAbstract:
This paper presents the usage of artificial neural networks (NNs) in bicycle route planning. This research aimed to check the possibility of NNs to transfer human expertise in bicycle route design by training the NN on an already established set of bicycle routes and then using the trained NN to design the routes on the novel area. We created two NNs capable of choosing the best route among the given road network by training them on two different areas. The bicycle routes produced by NNs were the same at best and had 75% overlap at the worst compared to those produced by human experts. Furthermore, the mean square error for all of our NN models varied from 0.015 and 0.081. We compared this new approach to the traditional multicriteria GIS (geographic information system) analysis (MA) that requires the human expert to define the bicycle route selection criteria. The benefit of using NN over the MA was that the NN directly transfers the human expertise to a model. In contrast, the MA needs the expert to select multiple criteria and adjust their weights carefully.
7
Hu, Yibo, Yuzhe Ou, Xujiang Zhao, Jin-Hee Cho, and Feng Chen. "Multidimensional Uncertainty-Aware Evidential Neural Networks." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 9 (May 18, 2021): 7815–22. http://dx.doi.org/10.1609/aaai.v35i9.16954.
Full textAbstract:
Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various application domains. However, NNs have not considered inherent uncertainty in data associated with the class probabilities where misclassification under uncertainty may easily introduce high risk in decision making in real-world contexts (e.g., misclassification of objects in roads leads to serious accidents). Unlike Bayesian NN that indirectly infer uncertainty through weight uncertainties, evidential NNs (ENNs) have been recently proposed to explicitly model the uncertainty of class probabilities and use them for classification tasks. An ENN offers the formulation of the predictions of NNs as subjective opinions and learns the function by collecting an amount of evidence that can form the subjective opinions by a deterministic NN from data. However, the ENN is trained as a black box without explicitly considering inherent uncertainty in data with their different root causes, such as vacuity (i.e., uncertainty due to a lack of evidence) or dissonance (i.e., uncertainty due to conflicting evidence). By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem. We took a hybrid approach that combines Wasserstein Generative Adversarial Network (WGAN) with ENNs to jointly train a model with prior knowledge of a certain class, which has high vacuity for OOD samples. Via extensive empirical experiments based on both synthetic and real-world datasets, we demonstrated that the estimation of uncertainty by WENN can significantly help distinguish OOD samples from boundary samples. WENN outperformed in OOD detection when compared with other competitive counterparts.
8
Krogh, Anders, and Søren Kamaric Riis. "Hidden Neural Networks." Neural Computation 11, no. 2 (February 1, 1999): 541–63. http://dx.doi.org/10.1162/089976699300016764.
Full textAbstract:
A general framework for hybrids of hidden Markov models (HMMs) and neural networks (NNs) called hidden neural networks (HNNs) is described. The article begins by reviewing standard HMMs and estimation by conditional maximum likelihood, which is used by the HNN. In the HNN, the usual HMM probability parameters are replaced by the outputs of state-specific neural networks. As opposed to many other hybrids, the HNN is normalized globally and therefore has a valid probabilistic interpretation. All parameters in the HNN are estimated simultaneously according to the discriminative conditional maximum likelihood criterion. The HNN can be viewed as an undirected probabilistic independence network (a graphical model), where the neural networks provide a compact representation of the clique functions. An evaluation of the HNN on the task of recognizing broad phoneme classes in the TIMIT database shows clear performance gains compared to standard HMMs tested on the same task.
9
DeVore, Ronald, Boris Hanin, and Guergana Petrova. "Neural network approximation." Acta Numerica 30 (May 2021): 327–444. http://dx.doi.org/10.1017/s0962492921000052.
Full textAbstract:
Neural networks (NNs) are the method of choice for building learning algorithms. They are now being investigated for other numerical tasks such as solving high-dimensional partial differential equations. Their popularity stems from their empirical success on several challenging learning problems (computer chess/Go, autonomous navigation, face recognition). However, most scholars agree that a convincing theoretical explanation for this success is still lacking. Since these applications revolve around approximating an unknown function from data observations, part of the answer must involve the ability of NNs to produce accurate approximations.This article surveys the known approximation properties of the outputs of NNs with the aim of uncovering the properties that are not present in the more traditional methods of approximation used in numerical analysis, such as approximations using polynomials, wavelets, rational functions and splines. Comparisons are made with traditional approximation methods from the viewpoint of rate distortion, i.e. error versus the number of parameters used to create the approximant. Another major component in the analysis of numerical approximation is the computational time needed to construct the approximation, and this in turn is intimately connected with the stability of the approximation algorithm. So the stability of numerical approximation using NNs is a large part of the analysis put forward.The survey, for the most part, is concerned with NNs using the popular ReLU activation function. In this case the outputs of the NNs are piecewise linear functions on rather complicated partitions of the domain of f into cells that are convex polytopes. When the architecture of the NN is fixed and the parameters are allowed to vary, the set of output functions of the NN is a parametrized nonlinear manifold. It is shown that this manifold has certain space-filling properties leading to an increased ability to approximate (better rate distortion) but at the expense of numerical stability. The space filling creates the challenge to the numerical method of finding best or good parameter choices when trying to approximate.
10
Wongsathan, Rati, and Pasit Pothong. "Heart Disease Classification Using Artificial Neural Networks." Applied Mechanics and Materials 781 (August 2015): 624–27. http://dx.doi.org/10.4028/www.scientific.net/amm.781.624.
Full textAbstract:
Neural Networks (NNs) has emerged as an importance tool for classification in the field of decision making. The main objective of this work is to design the structure and select the optimized parameter in the neural networks to implement the heart disease classifier. Three types of neural networks, i.e. Multi-layered Perceptron Neural Network (MLP-NN), Radial Basis Function Neural Networks (RBF-NN), and Generalized Regression Neural Network (GR-NN) have been used to test the performance of heart disease classification. The classification accuracy obtained by RBFNN gave a very high performance than MLP-NN and GR-NN respectively. The performance of accuracy is very promising compared with the previously reported another type of neural networks.
11
Krarti, M., J. F. Kreider, D. Cohen, and P. Curtiss. "Estimation of Energy Savings for Building Retrofits Using Neural Networks." Journal of Solar Energy Engineering 120, no. 3 (August 1, 1998): 211–16. http://dx.doi.org/10.1115/1.2888071.
Full textAbstract:
This paper overviews some applications of neural networks (NNs) to estimate energy and demand savings from retrofits of commercial buildings. First, a brief background information on NNs is provided. Then, three specific case studies are described to illustrate how and when NNs can be used successfully to determine energy savings due to the implementation of various energy conservation measures in existing commercial buildings.
12
Tebabal, Ambelu, Baylie Damtie, and Melessew Nigussie. "Temporal solar irradiance variability analysis using neural networks." Proceedings of the International Astronomical Union 11, S320 (August 2015): 333–38. http://dx.doi.org/10.1017/s1743921316000296.
Full textAbstract:
AbstractA feed-forward neural network which can account for nonlinear relationship was used to model total solar irradiance (TSI). A single layer feed-forward neural network with Levenberg-marquardt back-propagation algorithm have been implemented for modeling daily total solar irradiance from daily photometric sunspot index, and core-to-wing ratio of Mg II index data. In order to obtain the optimum neural network for TSI modeling, the root mean square error (RMSE) and mean absolute error (MAE) have been taken into account. The modeled and measured TSI have the correlation coefficient of about R=0.97. The neural networks (NNs) model output indicates that reconstructed TSI from solar proxies (photometric sunspot index and Mg II) can explain 94% of the variance of TSI. This modeled TSI using NNs further strengthens the view that surface magnetism indeed plays a dominant role in modulating solar irradiance.
13
Thorbjarnarson, Tómas, and Neil Yorke-Smith. "Optimal training of integer-valued neural networks with mixed integer programming." PLOS ONE 18, no. 2 (February 1, 2023): e0261029. http://dx.doi.org/10.1371/journal.pone.0261029.
Full textAbstract:
Recent work has shown potential in using Mixed Integer Programming (MIP) solvers to optimize certain aspects of neural networks (NNs). However the intriguing approach of training NNs with MIP solvers is under-explored. State-of-the-art-methods to train NNs are typically gradient-based and require significant data, computation on GPUs, and extensive hyper-parameter tuning. In contrast, training with MIP solvers does not require GPUs or heavy hyper-parameter tuning, but currently cannot handle anything but small amounts of data. This article builds on recent advances that train binarized NNs using MIP solvers. We go beyond current work by formulating new MIP models which improve training efficiency and which can train the important class of integer-valued neural networks (INNs). We provide two novel methods to further the potential significance of using MIP to train NNs. The first method optimizes the number of neurons in the NN while training. This reduces the need for deciding on network architecture before training. The second method addresses the amount of training data which MIP can feasibly handle: we provide a batch training method that dramatically increases the amount of data that MIP solvers can use to train. We thus provide a promising step towards using much more data than before when training NNs using MIP models. Experimental results on two real-world data-limited datasets demonstrate that our approach strongly outperforms the previous state of the art in training NN with MIP, in terms of accuracy, training time and amount of data. Our methodology is proficient at training NNs when minimal training data is available, and at training with minimal memory requirements—which is potentially valuable for deploying to low-memory devices.
14
Ko, Chien-Ho. "Predicting Subcontractor Performance Using Web-Based Evolutionary Fuzzy Neural Networks." Scientific World Journal 2013 (2013): 1–9. http://dx.doi.org/10.1155/2013/729525.
Full textAbstract:
Subcontractor performance directly affects project success. The use of inappropriate subcontractors may result in individual work delays, cost overruns, and quality defects throughout the project. This study develops web-based Evolutionary Fuzzy Neural Networks (EFNNs) to predict subcontractor performance. EFNNs are a fusion of Genetic Algorithms (GAs), Fuzzy Logic (FL), and Neural Networks (NNs). FL is primarily used to mimic high level of decision-making processes and deal with uncertainty in the construction industry. NNs are used to identify the association between previous performance and future status when predicting subcontractor performance. GAs are optimizing parameters required in FL and NNs. EFNNs encode FL and NNs using floating numbers to shorten the length of a string. A multi-cut-point crossover operator is used to explore the parameter and retain solution legality. Finally, the applicability of the proposed EFNNs is validated using real subcontractors. The EFNNs are evolved using 22 historical patterns and tested using 12 unseen cases. Application results show that the proposed EFNNs surpass FL and NNs in predicting subcontractor performance. The proposed approach improves prediction accuracy and reduces the effort required to predict subcontractor performance, providing field operators with web-based remote access to a reliable, scientific prediction mechanism.
15
Thoiyab, N. Mohamed, P. Muruganantham, Grienggrai Rajchakit, Nallappan Gunasekaran, Bundit Unyong, Usa Humphries, Pramet Kaewmesri, and Chee Peng Lim. "Global Stability Analysis of Neural Networks with Constant Time Delay via Frobenius Norm." Mathematical Problems in Engineering 2020 (October 12, 2020): 1–14. http://dx.doi.org/10.1155/2020/4321312.
Full textAbstract:
This paper deals with the global asymptotic robust stability (GARS) of neural networks (NNs) with constant time delay via Frobenius norm. The Frobenius norm result has been utilized to find a new sufficient condition for the existence, uniqueness, and GARS of equilibrium point of the NNs. Some suitable Lyapunov functional and the slope bounded functions have been employed to find the new sufficient condition for GARS of NNs. Finally, we give some comparative study of numerical examples for explaining the advantageous of the proposed result along with the existing GARS results in terms of network parameters.
16
Naz, Sidra, Muhammad Asif Zahoor Raja, Ammara Mehmood, and Aneela Zameer Jaafery. "Intelligent Predictive Solution Dynamics for Dahl Hysteresis Model of Piezoelectric Actuator." Micromachines 13, no. 12 (December 12, 2022): 2205. http://dx.doi.org/10.3390/mi13122205.
Full textAbstract:
Piezoelectric actuated models are promising high-performance precision positioning devices used for broad applications in the field of precision machines and nano/micro manufacturing. Piezoelectric actuators involve a nonlinear complex hysteresis that may cause degradation in performance. These hysteresis effects of piezoelectric actuators are mathematically represented as a second-order system using the Dahl hysteresis model. In this paper, artificial intelligence-based neurocomputing feedforward and backpropagation networks of the Levenberg–Marquardt method (LMM-NNs) and Bayesian Regularization method (BRM-NNs) are exploited to examine the numerical behavior of the Dahl hysteresis model representing a piezoelectric actuator, and the Adams numerical scheme is used to create datasets for various cases. The generated datasets were used as input target values to the neural network to obtain approximated solutions and optimize the values by using backpropagation neural networks of LMM-NNs and BRM-NNs. The performance analysis of LMM-NNs and BRM-NNs of the Dahl hysteresis model of the piezoelectric actuator is validated through convergence curves and accuracy measures via mean squared error and regression analysis.
17
AUDA, GASSER, and MOHAMED KAMEL. "MODULAR NEURAL NETWORKS: A SURVEY." International Journal of Neural Systems 09, no. 02 (April 1999): 129–51. http://dx.doi.org/10.1142/s0129065799000125.
Full textAbstract:
Modular Neural Networks (MNNs) is a rapidly growing field in artificial Neural Networks (NNs) research. This paper surveys the different motivations for creating MNNs: biological, psychological, hardware, and computational. Then, the general stages of MNN design are outlined and surveyed as well, viz., task decomposition techniques, learning schemes and multi-module decision-making strategies. Advantages and disadvantages of the surveyed methods are pointed out, and an assessment with respect to practical potential is provided. Finally, some general recommendations for future designs are presented.
18
Gharamohammadi, Ali, Mohammad Reza Yousefi Darestani, Ali Taghavirashidizadeh, Ahmad Abbasi, and Arash Shokouhmand. "Electromagnetic Sensor to Detect Objects by Classification and Neural Networks." Sensor Letters 17, no. 9 (September 1, 2019): 710–15. http://dx.doi.org/10.1166/sl.2019.4134.
Full textAbstract:
The backscatter signal analysis, as the landmine material could vary, has to be as much advanced as possible. One major problem with the conventional methods is that they are not able to detect new plastic landmines. In the recent research, the classification techniques and neural networks (NNs) were exploited for detection. In NNs-based method, a network is trained based on the feature extracted from the data, which leads to landmine detection. Other conventional classification methods, attempts to classify the objects sharing common characteristics. In this letter, an algorithm is introduced based on classification, data reduction and neural networks. Indeed, this algorithm employs neural network and classification method, simultaneously. The simple methods using either neural network or classification separately usually suffer from high rate of risk. In this letter, a novel classifier is proposed such that the data is classified based on similarity. It will be shown that the similarity between signals in a class is more than 90%, which proves the method's efficiency. Moreover, the scattering parameter, having magnitude and phase parts, is used to create an algorithm with parallel process.
19
Xiong, Zhi Bin. "Credit Risk Prediction Study Based on Modified Particle Swarm Optimized Fuzzy Neural Networks." Advanced Materials Research 108-111 (May 2010): 1326–31. http://dx.doi.org/10.4028/www.scientific.net/amr.108-111.1326.
Full textAbstract:
In last decade, neural networks (NNs) have been proposed to predict credit risk because of their advantages of treating non-linear data with self-learning capability. However, the shortcoming of NNs is also significant due to a “black box” syndrome. Moreover, in many situations NNs more or less suffer from the slow convergence and occasionally involve in a local optimal solution, which strongly limited their applications in practice. To overcome NN’s drawbacks, this paper presents a hybrid system that merges fuzzy neural network and niche evolution particle swarm optimization into a comprehensive mode, named as niche evolution particle swarm optimization fuzzy neural network (NEPSO-FNN), the new model has been applied to credit risk prediction based on the data collected from a set of Chinese listed corporations, and the results indicate that the performance of the proposed model is much better than the one of NN model using the cross-validation approach.
20
Jwo, Dah-Jing, Tai-Shen Lee, and Ying-Wei Tseng. "ARMA Neural Networks for Predicting DGPS Pseudorange Correction." Journal of Navigation 57, no. 2 (April 21, 2004): 275–86. http://dx.doi.org/10.1017/s0373463304002656.
Full textAbstract:
In this paper, the Auto-Regressive Moving-Averaging (ARMA) neural networks (NNs) will be incorporated for predicting the differential Global Positioning System (DGPS) pseudorange correction (PRC) information. The neural network is employed to realize the time-varying ARMA implementation. Online training for real-time prediction of the PRC enhances the continuity of service on the differential correction signals and therefore improves the positioning accuracy. When the PRC signal is lost, the ARMA neural network predicted PRC would temporarily provide correction data with very good accuracy. Simulation is conducted for evaluating the ARMA NN based DGPS PRC prediction accuracy. A comparative performance study based on two types of ARMA neural networks, i.e. Back-propagation Neural Network (BPNN) and General Regression Neural Network (GRNN), will be provided.
21
Shimada, Yukiyasu, and Kazuhiko Suzuki. "Feasibility Study of Fault Diagnostics Using Multiple Neural Networks." Journal of Robotics and Mechatronics 11, no. 4 (August 20, 1999): 326–30. http://dx.doi.org/10.20965/jrm.1999.p0326.
Full textAbstract:
This paper presents a fault diagnostic system (FDS) using multiple neural networks for chemical plants. The fault propagation in progress is modeled by causal relationships between a fault tree (FT) and its minimal cut sets (MCSa). The measurement patterns required for training neural networks (NNs) are obtained from fault propagation model. The FDS has a circuit network and component networks. The circuit network can identify circuit malfunctions that include disturbances. The component networks can identify component malfunctions as root causes of process malfunction. We have constructed an on-line FDS by making use of proposed method and verified the effectiveness of it experimentally.
22
Evci, Utku, Yani Ioannou, Cem Keskin, and Yann Dauphin. "Gradient Flow in Sparse Neural Networks and How Lottery Tickets Win." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 6 (June 28, 2022): 6577–86. http://dx.doi.org/10.1609/aaai.v36i6.20611.
Full textAbstract:
Sparse Neural Networks (NNs) can match the generalization of dense NNs using a fraction of the compute/storage for inference, and have the potential to enable efficient training. However, naively training unstructured sparse NNs from random initialization results in significantly worse generalization, with the notable exceptions of Lottery Tickets (LTs) and Dynamic Sparse Training (DST). In this work, we attempt to answer: (1) why training unstructured sparse networks from random initialization performs poorly and; (2) what makes LTs and DST the exceptions? We show that sparse NNs have poor gradient flow at initialization and propose a modified initialization for unstructured connectivity. Furthermore, we find that DST methods significantly improve gradient flow during training over traditional sparse training methods. Finally, we show that LTs do not improve gradient flow, rather their success lies in re-learning the pruning solution they are derived from — however, this comes at the cost of learning novel solutions.
23
Demichev, Andrey, and Alexander Kryukov. "Interrelation of equivariant Gaussian processes and convolutional neural networks." Journal of Physics: Conference Series 2438, no. 1 (February 1, 2023): 012095. http://dx.doi.org/10.1088/1742-6596/2438/1/012095.
Full textAbstract:
Abstract Currently there exists rather promising new trend in machine leaning (ML) based on the relationship between neural networks (NN) and Gaussian processes (GP), including many related subtopics, e.g., signal propagation in NNs, theoretical derivation of learning curve for NNs, QFT methods in ML, etc. An important feature of convolutional neural networks (CNN) is their equivariance (consistency) with respect to the symmetry transformations of the input data. In this work we establish a relationship between the many-channel limit for CNNs equivariant with respect to two-dimensional Euclidean group with vector-valued neuron activations and the corresponding independently introduced equivariant Gaussian processes (GP).
24
Hussain, Maab. "A RADIAL BASIS NEURAL NETWORK CONTROLLER TO SOLVE CONGESTION IN WIRELESS SENSOR NETWORKS." Iraqi Journal for Computers and Informatics 44, no. 1 (June 30, 2018): 40–48. http://dx.doi.org/10.25195/ijci.v44i1.103.
Full textAbstract:
In multihop networks, such as the Internet and the Mobile Ad-hoc Networks, routing is one of the most importantissues that has an important effect on the network’s performance. This work explores the possibility of using the shortest path routingin wireless sensor network . An ideal routing algorithm should combat to find an perfect path for data that transmitted within anexact time. First an overview of shortest path algorithm is given. Then a congestion estimation algorithm based on multilayerperceptron neural networks (MLP-NNs) with sigmoid activation function, (Radial Basis Neural Network Congestion Controller(RBNNCC) )as a controller at the memory space of the base station node. The trained network model was used to estimate trafficcongestion along the selected route. A comparison study between the network with and without controller in terms of: trafficreceived to the base station, execution time, data lost, and memory utilization . The result clearly shows the effectiveness of RadialBasis Neural Network Congestion Controller (RBNNCC) in traffic congestion prediction and control.
25
Wang, Haide, Ji Zhou, Yizhao Wang, Jinlong Wei, Weiping Liu, Changyuan Yu, and Zhaohui Li. "Optimization Algorithms of Neural Networks for Traditional Time-Domain Equalizer in Optical Communications." Applied Sciences 9, no. 18 (September 18, 2019): 3907. http://dx.doi.org/10.3390/app9183907.
Full textAbstract:
Neural networks (NNs) have been successfully applied to channel equalization for optical communications. In optical fiber communications, the linear equalizer and the nonlinear equalizer with traditional structures might be more appropriate than NNs for performing real-time digital signal processing, owing to its much lower computational complexity. However, the optimization algorithms of NNs are useful in many optimization problems. In this paper, we propose and evaluate the tap estimation schemes for the equalizer with traditional structures in optical fiber communications using the optimization algorithms commonly used in the NNs. The experimental results show that adaptive moment estimation algorithm and batch gradient descent method perform well in the tap estimation of equalizer. In conclusion, the optimization algorithms of NNs are useful in the tap estimation of equalizer with traditional structures in optical communications.
26
Smith, Robert E., and H. Brown Cribbs. "Is a Learning Classifier System a Type of Neural Network?" Evolutionary Computation 2, no. 1 (March 1994): 19–36. http://dx.doi.org/10.1162/evco.1994.2.1.19.
Full textAbstract:
This paper suggests a simple analogy between learning classifier systems (LCSs) and neural networks (NNs). By clarifying the relationship between LCSs and NNs, the paper indicates how techniques from one can be utilized in the other. The paper points out that the primary distinguishing characteristic of the LCS is its use of a co-adaptive genetic algorithm (GA), where the end product of evolution is a diverse population of individuals that cooperate to perform useful computation. This stands in contrast to typical GA/NN schemes, where a population of networks is employed to evolve a single, optimized network. To fully illustrate the LCS/NN analogy used in this paper, an LCS-like NN is implemented and tested. The test is constructed to run parallel to a similar GA/NN study that did not employ a co-adaptive GA. The test illustrates the LCS/NN analogy and suggests an interesting new method for applying GAs in NNs. Final comments discuss extensions of this work and suggest how LCS and NN studies can further benefit each other.
27
Jin, Yang. "Wavelet Scattering and Neural Networks for Railhead Defect Identification." Materials 14, no. 8 (April 14, 2021): 1957. http://dx.doi.org/10.3390/ma14081957.
Full textAbstract:
Accurate and automatic railhead inspection is crucial for the operational safety of railway systems. Deep learning on visual images is effective in the automatic detection of railhead defects, but either intensive data requirements or ignoring defect sizes reduce its applicability. This paper developed a machine learning framework based on wavelet scattering networks (WSNs) and neural networks (NNs) for identifying railhead defects. WSNs are functionally equivalent to deep convolutional neural networks while containing no parameters, thus suitable for non-intensive datasets. NNs can restore location and size information. The publicly available rail surface discrete defects (RSDD) datasets were analyzed, including 67 Type-I railhead images acquired from express tracks and 128 Type-II images captured from ordinary/heavy haul tracks. The ultimate validation accuracy reached 99.80% and 99.44%, respectively. WSNs can extract implicit signal features, and the support vector machine classifier can improve the learning accuracy of NNs by over 6%. Three criteria, namely the precision, recall, and F-measure, were calculated for comparison with the literature. At the pixel level, the developed approach achieved three criteria of around 90%, outperforming former methods. At the defect level, the recall rates reached 100%, indicating all labeled defects were identified. The precision rates were around 75%, affected by the insignificant misidentified speckles (smaller than 20 pixels). Nonetheless, the developed learning framework was effective in identifying railhead defects.
28
Bajcsy, Peter, Nicholas J. Schaub, and Michael Majurski. "Designing Trojan Detectors in Neural Networks Using Interactive Simulations." Applied Sciences 11, no. 4 (February 20, 2021): 1865. http://dx.doi.org/10.3390/app11041865.
Full textAbstract:
This paper addresses the problem of designing trojan detectors in neural networks (NNs) using interactive simulations. Trojans in NNs are defined as triggers in inputs that cause misclassification of such inputs into a class (or classes) unintended by the design of a NN-based model. The goal of our work is to understand encodings of a variety of trojan types in fully connected layers of neural networks. Our approach is: (1) to simulate nine types of trojan embeddings into dot patterns; (2) to devise measurements of NN states; and (3) to design trojan detectors in NN-based classification models. The interactive simulations are built on top of TensorFlow Playground with in-memory storage of data and NN coefficients. The simulations provide analytical, visualization, and output operations performed on training datasets and NN architectures. The measurements of a NN include: (a) model inefficiency using modified Kullback–Liebler (KL) divergence from uniformly distributed states; and (b) model sensitivity to variables related to data and NNs. Using the KL divergence measurements at each NN layer and per each predicted class label, a trojan detector is devised to discriminate NN models with or without trojans. To document robustness of such a trojan detector with respect to NN architectures, dataset perturbations, and trojan types, several properties of the KL divergence measurement are presented.
29
ROUSSAKI, IOANNA, IOANNIS PAPAIOANNOU, and MILTIADES ANAGNOSTOU. "USING NEURAL NETWORKS FOR EARLY DETECTION OF UNSUCCESSFUL NEGOTIATION THREADS." International Journal on Artificial Intelligence Tools 20, no. 03 (June 2011): 457–87. http://dx.doi.org/10.1142/s0218213011000231.
Full textAbstract:
Building agents that negotiate on behalf of their owners aiming to maximise their utility is a quite challenging research field in the artificial intelligence domain. In this paper, such agents are enhanced with techniques based on neural networks (NNs) to predict their opponents' negotiation behaviour, thus achieving more profitable results and better resource utilization. The NNs are used to early detect the cases where agreements are not achievable, supporting the decision of the agents to withdraw or not from the negotiation threads. The designed NN-assisted negotiation strategies have been evaluated via extensive experiments and are proven to be very useful.
30
Zhou, Shajun, Hongbing Zeng, and Wei Wang. "An Overview of Stability Analysis of Neural Networks with Time-Delays." Nanoscience and Nanotechnology Letters 11, no. 9 (September 1, 2019): 1200–1212. http://dx.doi.org/10.1166/nnl.2019.3009.
Full textAbstract:
Stability of neural networks (NNs) with time-delays (TDs) has become a popular research object. The Lyapunov-Krasovskii functional (LKF) method is an extremely effective technique for the stability research of NNs with TDs. Three primary issues, including ideas for constructing a reasonable LKF method, investigation of methods to estimate the LKF's derivative and reviewing approaches for establishing the stability criteria are discussed in details. Finally, some challenges are proposed for the future research directions.
31
Thakur, Amey. "Neuro-Fuzzy: Artificial Neural Networks & Fuzzy Logic." International Journal for Research in Applied Science and Engineering Technology 9, no. 9 (September 30, 2021): 128–35. http://dx.doi.org/10.22214/ijraset.2021.37930.
Full textAbstract:
Abstract: Neuro Fuzzy is a hybrid system that combines Artificial Neural Networks with Fuzzy Logic. Provides a great deal of freedom when it comes to thinking. This phrase, on the other hand, is frequently used to describe a system that combines both approaches. There are two basic streams of neural network and fuzzy system study. Modelling several elements of the human brain (structure, reasoning, learning, perception, and so on) as well as artificial systems and data: pattern clustering and recognition, function approximation, system parameter estimate, and so on. In general, neural networks and fuzzy logic systems are parameterized nonlinear computing methods for numerical data processing (signals, images, stimuli). These algorithms can be integrated into dedicated hardware or implemented on a general-purpose computer. The network system acquires knowledge through a learning process. Internal parameters are used to store the learned information (weights). Keywords: Artificial Neural Networks (ANNs), Neural Networks (NNs), Fuzzy Logic (FL), Neuro-Fuzzy, Probability Reasoning, Soft Computing, Fuzzification, Defuzzification, Fuzzy Inference Systems, Membership Function.
32
Yan, Ji Hong, Chao Zhong Guo, Xing Wang, and De Bin Zhao. "A Data-Driven Neural Network Approach for Remaining Useful Life Prediction." Key Engineering Materials 450 (November 2010): 544–47. http://dx.doi.org/10.4028/www.scientific.net/kem.450.544.
Full textAbstract:
This paper proposed a neural network (NN) based remaining useful life (RUL) prediction approach. A new performance degradation index is designed using multi-feature fusion techniques to represent deterioration severities of facilities. Based on this indicator, back propagation neural networks are trained for RUL prediction, and average of the networks’ outputs is considered as the final RUL in order to overcome prediction errors caused by random initiations of NNs. Finally, an experiment is set up based on a Bently-RK4 rotor unbalance test bed to validate the neural network based life prediction models, experimental results illustrate the effectiveness of the methodology.
33
Huang, Kai, and Cheng Wei Zhong. "Lightweight Concrete Strength Prediction by BP-ANN." Advanced Materials Research 1090 (February 2015): 101–6. http://dx.doi.org/10.4028/www.scientific.net/amr.1090.101.
Full textAbstract:
The back propagation artificial neural networks (BP-ANN) use a resilient back-propagation algorithm and early stopping technique. By inputing the properties of geometries and material, NNs can predict the strength of lightweight concrete. An BP-ANN model based on feed-forward neural network is built, trained and tested using the available test data of 148 mix records collected from the technical literature. And the test results are compared and analyzed with experimental data . It shows that the strength of lightweight concrete obtained by the simplified model based on NNs are in good agreement with test results, and they are close to the experimental values. The NNs model can be used in the shear strength prediction and design for the strength of lightweight concrete.
34
Yang, Wenqiang, Li Xiao, Junjian Huang, and Jinyue Yang. "Fixed-Time Synchronization of Neural Networks Based on Quantized Intermittent Control for Image Protection." Mathematics 9, no. 23 (November 30, 2021): 3086. http://dx.doi.org/10.3390/math9233086.
Full textAbstract:
This paper considers the fixed-time synchronization (FIXTS) of neural networks (NNs) by using quantized intermittent control (QIC). Based on QIC, a fixed-time controller is designed to ensure that the NNs achieve synchronization in finite time. With this controller, the settling time can be estimated regardless of initial conditions. After ensuring that the system has stabilized through this strategy, it is suitable for image protection given the behavior of the system. Meanwhile, the encryption effect of the image depends on the encryption algorithm, and the quality of the decrypted image depends on the synchronization error of NNs. The numerical results show that the designed controller is effective and validate the practical application of FIXTS of NNs in image protection.
35
Sheng, Chunyang, Haixia Wang, Xiao Lu, Zhiguo Zhang, Wei Cui, and Yuxia Li. "Distributed Gaussian Granular Neural Networks Ensemble for Prediction Intervals Construction." Complexity 2019 (July 3, 2019): 1–17. http://dx.doi.org/10.1155/2019/2379584.
Full textAbstract:
To overcome the weakness of generic neural networks (NNs) ensemble for prediction intervals (PIs) construction, a novel Map-Reduce framework-based distributed NN ensemble consisting of several local Gaussian granular NN (GGNNs) is proposed in this study. Each local network is weighted according to its contribution to the ensemble model. The weighted coefficient is estimated by evaluating the performance of the constructed PIs from each local network. A new evaluation principle is reported with the consideration of the predicting indices. To estimate the modelling uncertainty and the data noise simultaneously, the Gaussian granular is introduced to the numeric NNs. The constructed PIs can then be calculated by the variance of output distribution of each local NN, i.e., the summation of the model uncertainty variance and the data noise variance. To verify the effectiveness of the proposed model, a series of prediction experiments, including two classical time series with additive noise and two industrial time series, are carried out here. The results indicate that the proposed distributed GGNNs ensemble exhibits a good performance for PIs construction.
36
PAPAIOANNOU, IOANNIS, IOANNA ROUSSAKI, and MILTIADES ANAGNOSTOU. "USING NEURAL NETWORKS TO MINIMIZE THE DURATION OF AUTOMATED NEGOTIATION THREADS FOR HYBRID OPPONENTS." Journal of Circuits, Systems and Computers 19, no. 01 (February 2010): 59–74. http://dx.doi.org/10.1142/s0218126610005998.
Full textAbstract:
Neural networks (NNs) provide an efficient tool that can be trained to estimate the value of output parameters given certain metrics. In this paper, NNs are used to enhance intelligent agents that negotiate on behalf of their owners aiming to maximize their utility. More specifically, NNs are exploited in order to predict the hybrid negotiation behavior of the agents' opponents, thus achieving more profitable results for the parties these agents represent. The NNs provide the means so that the agents can early detect the cases where agreements are not achievable, thus supporting their decision to withdraw or not from the negotiation threads. The designed NN-assisted negotiation strategies have been evaluated via extensive experiments and are proven to be very useful, as they manage to significantly reduce the average duration of the negotiation threads.
37
Fernandes, Fabiano A. N., Francisco E. Linhares-Junior, and Samuel J. M. Cartaxo. "Prediction of Fischer–Tropsch Synthesis Kinetic Parameters Using Neural Networks." Chemical Product and Process Modeling 9, no. 2 (December 1, 2014): 97–103. http://dx.doi.org/10.1515/cppm-2013-0048.
Full textAbstract:
Abstract The kinetic mechanism of the Fischer–Tropsch synthesis (FTS) is complex resembling a polymerization reaction. The kinetic rate constants for initiation, propagation and termination steps and the constants for the equilibrium reactions for methylene formation (in situ monomer) need to be estimated. A mathematical model for the FTS allows for simulating several operating conditions and determining the best operating conditions to produce a specific product distribution, so the kinetic parameters must be statistically valid. This work used neural networks (NNs) to estimate the FTS kinetic parameters, instead of using methods based on least squared error. The results show that NNs with three hidden layers were able to output good estimates of the kinetic parameters with less than 5% of deviation.
38
Suantai, Suthep, Zulqurnain Sabir, Muhammad Asif Zahoor Raja, and Watcharaporn Cholamjiak. "A Stochastic Bayesian Neural Network for the Mosquito Dispersal Mathematical System." Fractal and Fractional 6, no. 10 (October 16, 2022): 604. http://dx.doi.org/10.3390/fractalfract6100604.
Full textAbstract:
The objective of this study is to examine numerical evaluations of the mosquito dispersal mathematical system (MDMS) in a heterogeneous atmosphere through artificial intelligence (AI) techniques via Bayesian regularization neural networks (BSR-NNs). The MDMS is constructed with six classes, i.e., eggs, larvae, pupae, host, resting mosquito, and ovipositional site densities-based ODEs system. The computing BSR-NNs scheme is applied for three different performances using the data of training, testing and verification, which is divided as 75%, 15%, 10% with twelve hidden neurons. The result comparisons are provided to check the authenticity of the designed AI method portrayed by the BSR-NNs. The AI based BSR-NNs procedure is executed to reduce the mean square error (MSE) for the MDMS. The achieved performances are also presented to validate the efficiency of BSR-NNs scheme using the process of MSE, correlation, error histograms and regression.
39
Monge Sanz, B. M., and N. J. Medrano Marqués. "Total ozone time series analysis: a neural network model approach." Nonlinear Processes in Geophysics 11, no. 5/6 (December 16, 2004): 683–89. http://dx.doi.org/10.5194/npg-11-683-2004.
Full textAbstract:
Abstract. This work is focused on the application of neural network based models to the analysis of total ozone (TO) time series. Processes that affect total ozone are extremely non linear, especially at the considered European mid-latitudes. Artificial neural networks (ANNs) are intrinsically non-linear systems, hence they are expected to cope with TO series better than classical statistics do. Moreover, neural networks do not assume the stationarity of the data series so they are also able to follow time-changing situations among the implicated variables. These two features turn NNs into a promising tool to catch the interactions between atmospheric variables, and therefore to extract as much information as possible from the available data in order to make, for example, time series reconstructions or future predictions. Models based on NNs have also proved to be very suitable for the treatment of missing values within the data series. In this paper we present several models based on neural networks to fill the missing periods of data within a total ozone time series, and models able to reconstruct the data series. The results released by the ANNs have been compared with those obtained by using classical statistics methods, and better accuracy has been achieved with the non linear ANNs techniques. Different network structures and training strategies have been tested depending on the specific task to be accomplished.
40
Zhang, Chen, Qingxu Li, and Xue Cheng. "Text Sentiment Classification Based on Feature Fusion." Revue d'Intelligence Artificielle 34, no. 4 (September 30, 2020): 515–20. http://dx.doi.org/10.18280/ria.340418.
Full textAbstract:
The convolutional neural network (CNN) and long short-term memory (LSTM) network are adept at extracting local and global features, respectively. Both can achieve excellent classification effects. However, the CNN performs poorly in extracting the global contextual information of the text, while LSTM often overlooks the features hidden between words. For text sentiment classification, this paper combines the CNN with bidirectional LSTM (BiLSTM) into a parallel hybrid model called CNN_BiLSTM. Firstly, the CNN was adopted to extract the local features of the text quickly. Next, the BiLSTM was employed to obtain the global text features containing contextual semantics. After that, the features extracted by the two neural networks (NNs) were fused, and processed by Softmax classifier for text sentiment classification. To verify its performance, the CNN_BiLSTM was compared with single NNs like CNN and LSTM, as well as other deep learning (DL) NNs through experiments. The experimental results show that the proposed parallel hybrid model outperformed the contrastive methods in F1-score and accuracy. Therefore, our model can solve text sentiment classification tasks effectively, and boast better practical value than other NNs.
41
Liang, Yunyi, Zhiyong Cui, Yu Tian, Huimiao Chen, and Yinhai Wang. "A Deep Generative Adversarial Architecture for Network-Wide Spatial-Temporal Traffic-State Estimation." Transportation Research Record: Journal of the Transportation Research Board 2672, no. 45 (October 8, 2018): 87–105. http://dx.doi.org/10.1177/0361198118798737.
Full textAbstract:
This study proposes a deep generative adversarial architecture (GAA) for network-wide spatial-temporal traffic-state estimation. The GAA is able to combine traffic-flow theory with neural networks and thus improve the accuracy of traffic-state estimation. It consists of two Long Short-Term Memory Neural Networks (LSTM NNs) which capture correlation in time and space among traffic flow and traffic density. One of the LSTM NNs, called a discriminative network, aims to maximize the probability of assigning correct labels to both true traffic-state matrices (i.e., traffic flow and traffic density within a given spatial-temporal area) and the traffic-state matrices generated from the other neural network. The other LSTM NN, called a generative network, aims to generate traffic-state matrices which maximize the probability that the discriminative network assigns true labels to them. The two LSTM NNs are trained simultaneously such that the trained generative network can generate traffic matrices similar to those in the training data set. Given a traffic-state matrix with missing values, we use back-propagation on three defined loss functions to map the corrupted matrix to a latent space. The mapping vector is then passed through the pre-trained generative network to estimate the missing values of the corrupted matrix. The proposed GAA is compared with the existing Bayesian network approach on loop detector data collected from Seattle, Washington and that collected from San Diego, California. Experimental results indicate that the GAA can achieve higher accuracy in traffic-state estimation than the Bayesian network approach.
42
Hou, Liyuan, Hong Zhu, Shouming Zhong, Yong Zeng, and Lin Shi. "State Estimation for Discrete-Time Stochastic Neural Networks with Mixed Delays." Journal of Applied Mathematics 2014 (2014): 1–14. http://dx.doi.org/10.1155/2014/209486.
Full textAbstract:
This paper investigates the analysis problem for stability of discrete-time neural networks (NNs) with discrete- and distribute-time delay. Stability theory and a linear matrix inequality (LMI) approach are developed to establish sufficient conditions for the NNs to be globally asymptotically stable and to design a state estimator for the discrete-time neural networks. Both the discrete delay and distribute delays employ decomposing the delay interval approach, and the Lyapunov-Krasovskii functionals (LKFs) are constructed on these intervals, such that a new stability criterion is proposed in terms of linear matrix inequalities (LMIs). Numerical examples are given to demonstrate the effectiveness of the proposed method and the applicability of the proposed method.
43
Heller, Daniel, Patrick Ferber, Julian Bitterwolf, Matthias Hein, and Jörg Hoffmann. "Neural Network Heuristic Functions: Taking Confidence into Account." Proceedings of the International Symposium on Combinatorial Search 15, no. 1 (July 17, 2022): 223–28. http://dx.doi.org/10.1609/socs.v15i1.21771.
Full textAbstract:
Neural networks (NN) are increasingly investigated in AI Planning, and are used successfully to learn heuristic functions. NNs commonly not only predict a value, but also output a confidence in this prediction. From the perspective of heuristic search with NN heuristics, it is a natural idea to take this into account, e.g. falling back to a standard heuristic where confidence is low. We contribute an empirical study of this idea. We design search methods which prune nodes, or switch between search queues, based on the confidence of NNs. We furthermore explore the possibility of out-of-distribution (OOD) training, which tries to reduce the overconfidence of NNs on inputs different to the training distribution. In experiments on IPC benchmarks, we find that our search methods improve coverage over standard methods, and that OOD training has the desired effect in terms of prediction accuracy and confidence, though its impact on search seems marginal.
44
Park, Namuk, Taekyu Lee, and Songkuk Kim. "Vector Quantized Bayesian Neural Network Inference for Data Streams." Proceedings of the AAAI Conference on Artificial Intelligence 35, no. 10 (May 18, 2021): 9322–30. http://dx.doi.org/10.1609/aaai.v35i10.17124.
Full textAbstract:
Bayesian neural networks (BNN) can estimate the uncertainty in predictions, as opposed to non-Bayesian neural networks (NNs). However, BNNs have been far less widely used than non-Bayesian NNs in practice since they need iterative NN executions to predict a result for one data, and it gives rise to prohibitive computational cost. This computational burden is a critical problem when processing data streams with low-latency. To address this problem, we propose a novel model VQ-BNN, which approximates BNN inference for data streams. In order to reduce the computational burden, VQ-BNN inference predicts NN only once and compensates the result with previously memorized predictions. To be specific, VQ-BNN inference for data streams is given by temporal exponential smoothing of recent predictions. The computational cost of this model is almost the same as that of non-Bayesian NNs. Experiments including semantic segmentation on real-world data show that this model performs significantly faster than BNNs while estimating predictive results comparable to or superior to the results of BNNs.
45
Anagnostopoulos, Ioannis, and Anas Rizeq. "Confining value from neural networks." Managerial Finance 45, no. 10/11 (October 14, 2019): 1433–57. http://dx.doi.org/10.1108/mf-12-2017-0523.
Full textAbstract:
Purpose This study provides valuable insights to managers aiming to increase the effectiveness of their diversification and growth portfolios. The purpose of this paper is to examine the value of utilizing a neural networks (NNs) approach using mergers and acquisition (M&A) data confined in the US technology domain. Design/methodology/approach Using data from Bloomberg for the period 2000–2016, the results confirm that an NN approach provides more explanation between financial variables in the model than a traditional regression model where the NN approach of this study is then compared with linear classifier, logistic regression. The empirical results show that NN is a promising method of evaluating M&A takeover targets in terms of their predictive accuracy and adaptability. Findings The findings emphasize the value alternative methodologies provide in high-technology industries in order to achieve the screening and explorative performance objectives, given the technological complexity, market uncertainty and the divergent skill sets required for breakthrough innovations in these sectors. Research limitations/implications NN methods do not provide for a fuller analysis of significance for each of the autonomous variables in the model as traditional regression methods do. The generalization breadth of this study is limited within a specific sector (technology) in a specific country (USA) covering a specific period (2000–2016). Practical implications Investors value firms before investing in them to identify their true stock price; yet, technology firms pose a great valuation challenge to investors and analysts alike as the latest information technology stock price bubbles, Silicon Valley and as the recent stratospheric rise of financial technology companies have also demonstrated. Social implications Numerous studies have shown that M&As are more often than not destroy value rather than create it. More than 50 percent of all M&As lead to a decline in relative total shareholder return after one year. Hence, effective target identification must be built on the foundation of a credible strategy that identifies the most promising market segments for growth, assesses whether organic or acquisitive growth is the best way forward and defines the commercial and financial hurdles for potential deals. Originality/value Technology firm value is directly dependent on growth, consequently most of the value will originate from future customers or products not from current assets that makes it challenging for investors to measure a firm’s beta (risk) where the value of a technology is only known after its commercialization to the market. A differentiated methodological approach used is the use of NNs, machine learning and data mining to predict bankruptcy or takeover targets.
46
Nagesh, A. "New Feature Vectors for Language Identification Using Deep Neural Networks." International Journal of Emerging Research in Management and Technology 6, no. 9 (June 24, 2018): 157. http://dx.doi.org/10.23956/ijermt.v6i9.103.
Full textAbstract:
The impressive performance of neural networks (NNs) for automatic speech recognition has motivated us to use for language identification (LID). In this paper, a new features based language identification system using neural network is presented. The new feature vectors are extracted based on the principle the frequency of occurrence phonemes is different among the languages. In this new form of feature vectors, the feature vectors are represented as a probability vector instead of scalar value. Because of this these new form of feature vectors, the DNN classifier classify the languages under consideration accurately.
47
Horng, Jenq-Ruey, Ming-Shyan Wang, Tai-Rung Lai, and Sergiu Berinde. "A neural observer for sensorless speed control of servomotors." Engineering Computations 31, no. 8 (October 28, 2014): 1668–78. http://dx.doi.org/10.1108/ec-11-2012-0289.
Full textAbstract:
Purpose – Extensive efforts have been conducted on the elimination of position sensors in servomotor control. The purpose of this paper is to aim at estimating the servomotor speed without using position sensors and the knowledge of its parameters by artificial neural networks (ANNs). Design/methodology/approach – A neural speed observer based on the Elman neural network (NN) structure takes only motor voltages and currents as inputs. Findings – After offline NNs training, the observer is incorporated into a DSP-based drive and sensorless control is achieved. Research limitations/implications – Future work will consider to reduce the computation time for NNs training and to adaptively tune parameters on line. Practical implications – The experimental results of the proposed method are presented to show the effectiveness. Originality/value – This paper achieves sensorless servomotor control by ANNs which are seldom studied.
48
Jiang, Yiming, Chenguang Yang, Jing Na, Guang Li, Yanan Li, and Junpei Zhong. "A Brief Review of Neural Networks Based Learning and Control and Their Applications for Robots." Complexity 2017 (2017): 1–14. http://dx.doi.org/10.1155/2017/1895897.
Full textAbstract:
As an imitation of the biological nervous systems, neural networks (NNs), which have been characterized as powerful learning tools, are employed in a wide range of applications, such as control of complex nonlinear systems, optimization, system identification, and patterns recognition. This article aims to bring a brief review of the state-of-the-art NNs for the complex nonlinear systems by summarizing recent progress of NNs in both theory and practical applications. Specifically, this survey also reviews a number of NN based robot control algorithms, including NN based manipulator control, NN based human-robot interaction, and NN based cognitive control.
49
Tao, Zhou, Hou Muzhou, and Liu Chunhui. "Forecasting stock index with multi-objective optimization model based on optimized neural network architecture avoiding overfitting." Computer Science and Information Systems 15, no. 1 (2018): 211–36. http://dx.doi.org/10.2298/csis170125042t.
Full textAbstract:
In this paper, the stock index time series forecasting using optimal neural networks with optimal architecture avoiding overfitting is studied. The problem of neural network architecture selection is a central problem in the application of neural network computation. After analyzing the reasons for overfitting and instability of neural networks, in order to find the optimal NNs (neural networks) architecture, we consider minimizing three objective indexes: training and testing root mean square error (RMSE) and testing error variance (TEV). Then we built a multi-objective optimization model, then converted it to single objective optimization model and proved the existence and uniqueness theorem of optimal solution. After determining the searching interval, a Multiobjective Optimization Algorithm for Optimized Neural Network Architecture Avoiding Overfitting (ONNAAO) is constructed to solve above model and forecast the time series. Some experiments with several different datasets are taken for training and forecasting. And some performance such as training time, testing RMSE and neurons, has been compared with the traditional algorithm (AR, ARMA, ordinary BP, SVM) through many numerical experiments, which fully verified the superiority, correctness and validity of the theory.
50
D’Ambrosio, Andrea, Enrico Schiassi, Fabio Curti, and Roberto Furfaro. "Pontryagin Neural Networks with Functional Interpolation for Optimal Intercept Problems." Mathematics 9, no. 9 (April 28, 2021): 996. http://dx.doi.org/10.3390/math9090996.
Full textAbstract:
In this work, we introduce Pontryagin Neural Networks (PoNNs) and employ them to learn the optimal control actions for unconstrained and constrained optimal intercept problems. PoNNs represent a particular family of Physics-Informed Neural Networks (PINNs) specifically designed for tackling optimal control problems via the Pontryagin Minimum Principle (PMP) application (e.g., indirect method). The PMP provides first-order necessary optimality conditions, which result in a Two-Point Boundary Value Problem (TPBVP). More precisely, PoNNs learn the optimal control actions from the unknown solutions of the arising TPBVP, modeling them with Neural Networks (NNs). The characteristic feature of PoNNs is the use of PINNs combined with a functional interpolation technique, named the Theory of Functional Connections (TFC), which forms the so-called PINN-TFC based frameworks. According to these frameworks, the unknown solutions are modeled via the TFC’s constrained expressions using NNs as free functions. The results show that PoNNs can be successfully applied to learn optimal controls for the class of optimal intercept problems considered in this paper.