Search results for "Artificial neural network"

showing 10 items of 694 documents

Diffusive neural network

2002

Abstract A non-connectionist model of a neuronal network based on passive diffusion of neurotransmitters is presented as an alternative to hard-wired artificial neural networks. Classic thermodynamical approach shows that the diffusive network is capable of exhibiting asymptotic stability and a dynamics resembling that of a chaotic system. Basic computational capabilities of the net are discussed based on the equivalence with a Turing machine. The model offers a way to represent mass-sustained brain functions in terms of recurrent behaviors in the phase space.

Theoretical computer scienceQuantitative Biology::Neurons and CognitionArtificial neural networkComputer scienceCognitive NeuroscienceChaoticTopologyComputer Science ApplicationsTuring machinesymbols.namesakeRecurrent neural networkExponential stabilityArtificial IntelligencePhase spacesymbolsBiological neural networkStochastic neural networkNeurocomputing
researchProduct

The computational power of continuous time neural networks

1997

We investigate the computational power of continuous-time neural networks with Hopfield-type units. We prove that polynomial-size networks with saturated-linear response functions are at least as powerful as polynomially space-bounded Turing machines.

TheoryofComputation_COMPUTATIONBYABSTRACTDEVICESQuantitative Biology::Neurons and CognitionComputational complexity theoryArtificial neural networkComputer sciencebusiness.industryComputer Science::Neural and Evolutionary ComputationNSPACEComputational resourcePower (physics)Turing machinesymbols.namesakeCellular neural networksymbolsArtificial intelligenceTypes of artificial neural networksbusiness
researchProduct

Comparative study of modelling the thermal efficiency of a novel straight through evacuated tube collector with MLR, SVR, BP and RBF methods

2021

Abstract Data-based methods are useful for accurate modelling of solar thermal systems. In this work, several artificial neural network (ANN) techniques are proposed to predict the thermal performance of an all-glass straight through evacuated tube solar collector. These are compared to support vector regression analysis. Extensive experimental data sets were collected for training the ANN models. Solar radiation intensity, ambient temperature, wind speed, mass flow rate and collector inlet temperature were selected as the input layer to predict the thermal efficiency of the solar collector. The prediction precision of the ANN models was compared to the multiple linear regression and suppor…

Thermal efficiencyArtificial neural networkRenewable Energy Sustainability and the Environment020209 energyEnergy Engineering and Power Technology02 engineering and technologyMechanicsWind speedBackpropagationSupport vector machine020401 chemical engineeringThermalLinear regression0202 electrical engineering electronic engineering information engineeringMass flow rateEnvironmental science0204 chemical engineeringSustainable Energy Technologies and Assessments
researchProduct

Neural Network Based Finite-Time Stabilization for Discrete-Time Markov Jump Nonlinear Systems with Time Delays

2013

Published version of an article in the journal: Abstract and Applied Analysis. Also available from the publisher at: http://dx.doi.org/10.1155/2013/359265 Open Access This paper deals with the finite-time stabilization problem for discrete-time Markov jump nonlinear systems with time delays and norm-bounded exogenous disturbance. The nonlinearities in different jump modes are parameterized by neural networks. Subsequently, a linear difference inclusion state space representation for a class of neural networks is established. Based on this, sufficient conditions are derived in terms of linear matrix inequalities to guarantee stochastic finite-time boundedness and stochastic finite-time stabi…

Time delaysArticle SubjectState-space representationArtificial neural networklcsh:MathematicsApplied MathematicsParameterized complexitylcsh:QA1-939VDP::Mathematics and natural science: 400::Mathematics: 410::Analysis: 411Nonlinear systemDiscrete time and continuous timeControl theoryJumpAnalysisMathematicsMarkov jumpAbstract and Applied Analysis
researchProduct

Synchronization of Uncertain Neural Networks with H8 Performance and Mixed Time-Delays

2011

An exponential H8 synchronization method is addressed for a class of uncertain master and slave neural networks with mixed time-delays, where the mixed delays comprise different neutral, discrete and distributed time-delays. An appropriate discretized Lyapunov-Krasovskii functional and some free weighting matrices are utilized to establish some delay-dependent sufficient conditions for designing a delayed state-feedback control as a synchronization law in terms of linear matrix inequalities under less restrictive conditions. The controller guarantees the exponential H8 synchronization of the two coupled master and slave neural networks regardless of their initial states. Numerical simulatio…

Time delaysArtificial neural networkComputer scienceControl theorySynchronization (computer science)
researchProduct

Notice of Violation of IEEE Publication Principles: New Delay-Dependent Exponential $H_{\infty}$ Synchronization for Uncertain Neural Networks With M…

2010

This paper establishes an exponential H infin synchronization method for a class of uncertain master and slave neural networks (MSNNs) with mixed time delays, where the mixed delays comprise different neutral, discrete, and distributed time delays. The polytopic and the norm-bounded uncertainties are separately taken into consideration. An appropriate discretized Lyapunov-Krasovskii functional and some free-weighting matrices are utilized to establish some delay-dependent sufficient conditions for designing delayed state-feedback control as a synchronization law in terms of linear matrix inequalities under less restrictive conditions. The controller guarantees the exponential H infin synchr…

Time delaysDiscretizationArtificial neural networkGeneral MedicineLinear matrixSynchronizationComputer Science ApplicationsExponential functionHuman-Computer InteractionDelay dependentControl and Systems EngineeringControl theoryElectrical and Electronic EngineeringSoftwareInformation SystemsMathematicsIEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics)
researchProduct

Computer-Aided Diagnosis System with Backpropagation Artificial Neural Network—Improving Human Readers Performance

2016

This article presents the results of a study into possibility of artificial neural networks (ANNs) to classify cancer changes in mammographic images. Today’s Computer-Aided Detection (CAD) systems cannot detect 100 % of pathological changes. One of the properties of an ANN is generalized information —it can identify not only learned data but also data that is similar to training set. The combination of CAD and ANN could give better result and help radiologists to take the right decision.

Training setArtificial neural networkComputer sciencebusiness.industryComputer Science::Neural and Evolutionary ComputationPhysics::Medical PhysicsCADMachine learningcomputer.software_genreComputer aided detectionComputingMethodologies_PATTERNRECOGNITIONComputer-aided diagnosisArtificial intelligencebusinessartificial neural networks�mammographic imagescomputercomputer-aided detectionBackpropagation artificial neural network
researchProduct

Multilayer neural networks: an experimental evaluation of on-line training methods

2004

Artificial neural networks (ANN) are inspired by the structure of biological neural networks and their ability to integrate knowledge and learning. In ANN training, the objective is to minimize the error over the training set. The most popular method for training these networks is back propagation, a gradient descent technique. Other non-linear optimization methods such as conjugate directions set or conjugate gradient have also been used for this purpose. Recently, metaheuristics such as simulated annealing, genetic algorithms or tabu search have been also adapted to this context.There are situations in which the necessary training data are being generated in real time and, an extensive tr…

Training setGeneral Computer ScienceArtificial neural networkbusiness.industryComputer scienceComputer Science::Neural and Evolutionary ComputationMathematicsofComputing_NUMERICALANALYSISContext (language use)Management Science and Operations ResearchMachine learningcomputer.software_genreBackpropagationTabu searchModeling and SimulationConjugate gradient methodGenetic algorithmSimulated annealingArtificial intelligencebusinessGradient descentcomputerMetaheuristicComputers & Operations Research
researchProduct

Towards to deep neural network application with limited training data: synthesis of melanoma's diffuse reflectance spectral images

2019

The goal of our study is to train artificial neural networks (ANN) using multispectral images of melanoma. Since the number of multispectral images of melanomas is limited, we offer to synthesize them from multispectral images of benign skin lesions. We used the previously created melanoma diagnostic criterion p'. This criterion is calculated from multispectral images of skin lesions captured under 526nm, 663nm, and 964nm LED illumination. We synthesize these three images from multispectral images of nevus so that the p' map matches the melanoma criteria (the values in the lesion area is >1, respectively). Demonstrated results show that by transforming multispectral images of benign nevus i…

Training setLed illuminationArtificial neural networkbusiness.industryComputer scienceMelanomaMultispectral imagePattern recognitionmedicine.diseasemedicineNevusBenign nevusArtificial intelligenceSkin cancerbusinessDiffuse Optical Spectroscopy and Imaging VII
researchProduct

Pairwise Learning to Rank by Neural Networks Revisited: Reconstruction, Theoretical Analysis and Practical Performance

2020

We present a pairwise learning to rank approach based on a neural net, called DirectRanker, that generalizes the RankNet architecture. We show mathematically that our model is reflexive, antisymmetric, and transitive allowing for simplified training and improved performance. Experimental results on the LETOR MSLR-WEB10K, MQ2007 and MQ2008 datasets show that our model outperforms numerous state-of-the-art methods, while being inherently simpler in structure and using a pairwise approach only.

Transitive relationPairwise learningTheoretical computer scienceArtificial neural networkAntisymmetric relationComputer scienceRank (computer programming)Structure (category theory)Pairwise comparisonLearning to rank
researchProduct