Search results for "Perceptron"

showing 10 items of 89 documents

Assigning discounts in a marketing campaign by using reinforcement learning and neural networks

2009

In this work, RL is used to find an optimal policy for a marketing campaign. Data show a complex characterization of state and action spaces. Two approaches are proposed to circumvent this problem. The first approach is based on the self-organizing map (SOM), which is used to aggregate states. The second approach uses a multilayer perceptron (MLP) to carry out a regression of the action-value function. The results indicate that both approaches can improve a targeted marketing campaign. Moreover, the SOM approach allows an intuitive interpretation of the results, and the MLP approach yields robust results with generalization capabilities.

Artificial neural networkComputer scienceGeneralizationbusiness.industrymedia_common.quotation_subjectAggregate (data warehouse)General EngineeringMachine learningcomputer.software_genreComputer Science ApplicationsFunction approximationArtificial IntelligenceMultilayer perceptronReinforcement learningState (computer science)Artificial intelligenceFunction (engineering)businesscomputermedia_commonExpert Systems with Applications
researchProduct

Crane collision modelling using a neural network approach

2004

Abstract The objective of the present work is to find a Collision Detection algorithm to be used in the Virtual Reality crane simulator (UVSim®), developed by the Robotics Institute of the University of Valencia for the Port of Valencia. The method is applicable to box-shaped objects and is based on the relationship between the colliding object positions and their impact points. The tool chosen to solve the problem is a neural network, the multilayer perceptron, which adapts to the characteristics of the problem, namely, non-linearity, a large amount of data, and no a priori knowledge. The results achieved by the neural network are very satisfactory for the case of box-shaped objects. Furth…

Artificial neural networkComputer sciencebusiness.industryGeneral EngineeringRoboticsObject (computer science)CollisionComputer Science ApplicationsArtificial IntelligenceSimulació per ordinadorMultilayer perceptronXarxes neuronals (Informàtica)Collision detectionArtificial intelligencebusinessAlgorithmGantry craneExpert Systems with Applications
researchProduct

Efficient pruning of multilayer perceptrons using a fuzzy sigmoid activation function

2006

This Letter presents a simple and powerful pruning method for multilayer feed forward neural networks based on the fuzzy sigmoid activation function presented in [E. Soria, J. Martin, G. Camps, A. Serrano, J. Calpe, L. Gomez, A low-complexity fuzzy activation function for artificial neural networks, IEEE Trans. Neural Networks 14(6) (2003) 1576-1579]. Successful performance is obtained in standard function approximation and channel equalization problems. Pruning allows to reduce network complexity considerably, achieving a similar performance to that obtained by unpruned networks.

Artificial neural networkComputer sciencebusiness.industryTime delay neural networkCognitive NeuroscienceActivation functionRectifier (neural networks)PerceptronFuzzy logicComputer Science ApplicationsArtificial IntelligenceMultilayer perceptronFeedforward neural networkPruning (decision trees)Artificial intelligenceTypes of artificial neural networksbusinessNeurocomputing
researchProduct

Classical Training Methods

2006

This chapter reviews classical training methods for multilayer neural networks. These methods are widely used for classification and function modelling tasks. Nevertheless, they show a number of flaws or drawbacks that should be addressed in the development of such systems. They work by searching the minimum of an error function which defines the optimal behaviour of the neural network. Different standard problems are used to show the capabilities of these models; in particular, we have benchmarked the algorithms in a nonlinear classification problem and in three function modelling problems.

Artificial neural networkComputer sciencebusiness.industrymedia_common.quotation_subjectTraining methodsMachine learningcomputer.software_genreError functionDelta ruleMultilayer perceptronArtificial intelligenceNonlinear classificationbusinessFunction (engineering)computermedia_common
researchProduct

Deep learning strategies for automatic fault diagnosis in photovoltaic systems by thermographic images

2021

Abstract Losses of electricity production in photovoltaic systems are mainly caused by the presence of faults that affect the efficiency of the systems. The identification of any overheating in a photovoltaic module, through the thermographic non-destructive test, may be essential to maintain the correct functioning of the photovoltaic system quickly and cost-effectively, without interrupting its normal operation. This work proposes a system for the automatic classification of thermographic images using a convolutional neural network, developed via open-source libraries. To reduce image noise, various pre-processing strategies were evaluated, including normalization and homogenization of pi…

Artificial neural networkContextual image classificationRenewable Energy Sustainability and the EnvironmentComputer sciencebusiness.industry020209 energyDeep learningEnergy Engineering and Power TechnologyPattern recognitionSobel operatorAutomatic Fault recognition Convolutional Neural Network Photovoltaics TensorFlow Infrared Thermography02 engineering and technologyPerceptronConvolutional neural networkThresholdingThermographic inspectionFuel Technology020401 chemical engineeringNuclear Energy and Engineering0202 electrical engineering electronic engineering information engineeringArtificial intelligence0204 chemical engineeringbusinessEnergy Conversion and Management
researchProduct

Identification of the Parameters of Reduced Vector Preisach Model by Neural Networks

2008

This paper presents a methodology for identifying reduced vector Preisach model parameters by using neural networks. The neural network used is a multiplayer perceptron trained with the Levenberg-Marquadt training algorithm. The network is trained by some hysteresis data, which are generated by using reduced vector Preisach model with preassigned parameters. It is shown how a properly trained network is able to find the parameters needed to best fit a magnetization hysteresis curve.

Artificial neural networkEstimation theoryComputer sciencebusiness.industryDifferential equationComputer Science::Neural and Evolutionary ComputationPattern recognitionMagnetic hysteresisPerceptronMagnetic susceptibilityElectronic Optical and Magnetic MaterialsIdentification (information)MagnetizationHysteresisMultilayer perceptronArtificial intelligenceElectrical and Electronic EngineeringbusinessSaturation (magnetic)
researchProduct

A Memetic-Neural Approach to Discover Resources in P2P Networks

2008

This chapter proposes a neural network based approach for solving the resource discovery problem in Peer to Peer (P2P) networks and an Adaptive Global Local Memetic Algorithm (AGLMA) for performing in training of the neural network. The neural network, which is a multi-layer perceptron neural network, allows the P2P nodes to efficiently locate resources desired by the user. The necessity of testing the network in various working conditions, aiming to obtain a robust neural network, introduces noise in the objective function. The AGLMA is a memetic algorithm which employs two local search algorithms adaptively activated by an evolutionary framework. These local searchers, having different fe…

Artificial neural networkbusiness.industryProcess (engineering)Computer scienceComputer Science::Neural and Evolutionary ComputationComputational intelligencePeer-to-peercomputer.software_genrePerceptronMachine learningResource (project management)Memetic algorithmLocal search (optimization)Artificial intelligencebusinesscomputer
researchProduct

Short term wind speed prediction using Multi Layer Perceptron

2012

Among renewable energy sources wind energy is having an increasing influence on the supply of energy power. However wind energy is not a stationary power, depending on the fluctuations of the wind, so that is necessary to cope with these fluctuations that may cause problems the electricity grid stability. The ability to predict short-term wind speed and consequent production patterns becomes critical for the all the operators of wind energy. This paper studies several configurations of Artificial Neural Networks (ANN), a well-known tool able to estimate wind speed starting from measured data. The presented ANNs, t have been tested through data gathered in the area of Trapani (Sicily). Diffe…

Artificial neural networks Multi layer perceptron Feed forward network Forecasting Renewable energy Wind energy Wind speedSettore ING-IND/11 - Fisica Tecnica Ambientale
researchProduct

Air pollution in European countries and life expectancy—modelling with the use of neural network

2019

Abstract The present paper discusses a novel methodology based on neural network to determine air pollutants’ correlation with life expectancy in European countries. The models were developed using historical data from the period 1992–2016, for a set of 20 European countries. The subject of the analysis included the input variables of the following air pollutants: sulphur oxides, nitrogen oxides, carbon monoxide, particulate matters, polycyclic aromatic hydrocarbons and non-methane volatile organic compounds. Our main findings indicate that all the variables significantly affect life expectancy. Sensitivity of constructed neural networks to pollutants proved to be particularly important in …

Atmospheric Science010504 meteorology & atmospheric sciencesHealth Toxicology and Mutagenesisair pollutionAir pollution010501 environmental sciencesManagement Monitoring Policy and Lawmedicine.disease_cause01 natural sciencesmodellingmulti-layer perceptronAir pollutantsEconometricsmedicineNitrogen oxides0105 earth and related environmental sciencesPollutantArtificial neural networkpublic healthParticulatesPollutionVariable (computer science)Life expectancylife expectancyEnvironmental scienceartificial neural networkAir Quality Atmosphere and Health
researchProduct

A method for modeling the battery state of charge in wireless sensor networks

2015

In this paper we propose a method for obtaining an analytic model of the battery State-of-Charge (SoC) in wireless sensor nodes. The objective is to find simple models that can be used to estimate accurately the real battery state and consequently the node lifetime. Running the model in the network nodes, we can provide the motes with the required information to implement applications that can be considered as battery-aware. The proposed methodology reduces the computational complexity of the model avoiding complicated electrochemical simulations and treating the battery as an unknown system with an output that can be predicted using simple mathematical models. At a first stage, during a se…

Battery (electricity)EngineeringEnergyMathematical modelbusiness.industryNode (networking)Real-time computingWireless sensor networksTECNOLOGIA ELECTRONICAKey distribution in wireless sensor networksMultilayer perceptronComputer Science::Networking and Internet ArchitectureElectronic engineeringMobile wireless sensor networkBatteries State-of-ChargeWirelessBatteries modelingElectrical and Electronic EngineeringbusinessInstrumentationWireless sensor network
researchProduct