Search results for "hierarchy"

showing 10 items of 312 documents

On the impact of forgetting on learning machines

1995

People tend not to have perfect memories when it comes to learning, or to anything else for that matter. Most formal studies of learning, however, assume a perfect memory. Some approaches have restricted the number of items that could be retained. We introduce a complexity theoretic accounting of memory utilization by learning machines. In our new model, memory is measured in bits as a function of the size of the input. There is a hierarchy of learnability based on increasing memory allotment. The lower bound results are proved using an unusual combination of pumping and mutual recursion theorem arguments. For technical reasons, it was necessary to consider two types of memory : long and sh…

Theoretical computer scienceActive learning (machine learning)Computer scienceSemi-supervised learningMutual recursionArtificial IntelligenceInstance-based learningHierarchyForgettingKolmogorov complexitybusiness.industryLearnabilityAlgorithmic learning theoryOnline machine learningInductive reasoningPumping lemma for regular languagesTerm (time)Computational learning theoryHardware and ArchitectureControl and Systems EngineeringArtificial intelligenceSequence learningbusinessSoftwareCognitive psychologyInformation SystemsJournal of the ACM
researchProduct

The Hierarchical Continuous Pursuit Learning Automation: A Novel Scheme for Environments With Large Numbers of Actions.

2019

Although the field of learning automata (LA) has made significant progress in the past four decades, the LA-based methods to tackle problems involving environments with a large number of actions is, in reality, relatively unresolved. The extension of the traditional LA to problems within this domain cannot be easily established when the number of actions is very large. This is because the dimensionality of the action probability vector is correspondingly large, and so, most components of the vector will soon have values that are smaller than the machine accuracy permits, implying that they will never be chosen . This paper presents a solution that extends the continuous pursuit paradigm to …

Theoretical computer scienceHierarchical learning automataHierarchy (mathematics)DiscretizationLearning automataComputer Networks and CommunicationsComputer scienceLarge action numbersPursuit learning automata02 engineering and technologyVDP::Matematikk og Naturvitenskap: 400::Informasjons- og kommunikasjonsvitenskap: 420Probability vectorLearning automataComputer Science ApplicationsAutomatonOperator (computer programming)Artificial Intelligence0202 electrical engineering electronic engineering information engineeringBenchmark (computing)Estimator-based learning automata020201 artificial intelligence & image processingVDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550SoftwareCurse of dimensionalityIEEE transactions on neural networks and learning systems
researchProduct

The Hierarchical Continuous Pursuit Learning Automation for Large Numbers of Actions

2018

Part 10: Learning - Intelligence; International audience; Although the field of Learning Automata (LA) has made significant progress in the last four decades, the LA-based methods to tackle problems involving environments with a large number of actions are, in reality, relatively unresolved. The extension of the traditional LA (fixed structure, variable structure, discretized, and pursuit) to problems within this domain cannot be easily established when the number of actions is very large. This is because the dimensionality of the action probability vector is correspondingly large, and consequently, most components of the vector will, after a relatively short time, have values that are smal…

Theoretical computer scienceHierarchical learning automataHierarchy (mathematics)Learning automataComputer sciencePursuit learning automataPursuit LALearning Automata02 engineering and technologyEstimator-based LAProbability vectorField (computer science)020202 computer hardware & architectureLA with large number of actionsVariable (computer science)Operator (computer programming)Learning Automata (LA)Action (philosophy)0202 electrical engineering electronic engineering information engineeringEstimator-based learning automata[INFO]Computer Science [cs]020201 artificial intelligence & image processingHierarchical LACurse of dimensionality
researchProduct

Diagrammatic approach to cellular automata and the emergence of form with inner structure

2018

We present a diagrammatic method to build up sophisticated cellular automata (CAs) as models of complex physical systems. The diagrams complement the mathematical approach to CA modeling, whose details are also presented here, and allow CAs in rule space to be classified according to their hierarchy of layers. Since the method is valid for any discrete operator and only depends on the alphabet size, the resulting conclusions, of general validity, apply to CAs in any dimension or order in time, arbitrary neighborhood ranges and topology. We provide several examples of the method, illustrating how it can be applied to the mathematical modeling of the emergence of order out of disorder. Specif…

Theoretical computer scienceStructure (category theory)Physical systemFOS: Physical sciencesPattern Formation and Solitons (nlin.PS)01 natural sciences010305 fluids & plasmasOperator (computer programming)0103 physical sciences010306 general physicsTopology (chemistry)Mathematical PhysicsMathematicsComplement (set theory)Numerical AnalysisHierarchy (mathematics)Applied MathematicsCellular Automata and Lattice Gases (nlin.CG)Mathematical Physics (math-ph)Nonlinear Sciences - Pattern Formation and SolitonsCellular automatonNonlinear Sciences - Adaptation and Self-Organizing SystemsDiagrammatic reasoningModeling and SimulationAlgorithmAdaptation and Self-Organizing Systems (nlin.AO)Nonlinear Sciences - Cellular Automata and Lattice Gases
researchProduct

Quantum Real - Time Turing Machine

2001

The principles of quantum computation differ from the principles of classical computation very much. Quantum analogues to the basic constructions of the classical computation theory, such as Turing machine or finite 1-way and 2-ways automata, do not generalize deterministic ones. Their capabilities are incomparable. The aim of this paper is to introduce a quantum counterpart for real - time Turing machine. The recognition of a special kind of language, that can't be recognized by a deterministic real - time Turing machine, is shown.

TheoryofComputation_COMPUTATIONBYABSTRACTDEVICESTheoretical computer scienceQuantum Turing machineDTIMEComputer scienceProbabilistic Turing machine2-EXPTIMESuper-recursive algorithmComputationDescription numberDSPACElaw.inventionsymbols.namesakeTuring machineTuring completenessNon-deterministic Turing machinelawAlgorithm characterizationsQuantumPSPACEQuantum computerFinite-state machineTuring machine examplesNSPACETheoryofComputation_GENERALAutomatonTheoryofComputation_MATHEMATICALLOGICANDFORMALLANGUAGESTuring reductionTheory of computationsymbolsUniversal Turing machineTime hierarchy theoremAlternating Turing machineComputer Science::Formal Languages and Automata TheoryRegister machine
researchProduct

Space-Efficient 1.5-Way Quantum Turing Machine

2001

1.5QTM is a sort of QTM (Quantum Turing Machine) where the head cannot move left (it can stay where it is and move right). For computations is used other - work tape. In this paper will be studied possibilities to economize work tape space more than the same deterministic Turing Machine can do (for some of the languages). As an example language (0i1i|i ≥ 0) is chosen, and is proved that this language could be recognized by deterministic Turing machine using log(i) cells on work tape , and 1.5QTM can recognize it using constant cells quantity.

TheoryofComputation_COMPUTATIONBYABSTRACTDEVICESTheoretical computer scienceQuantum Turing machineSuper-recursive algorithmComputer scienceProbabilistic Turing machineComputationDescription numberMultitape Turing machineDSPACElaw.inventionTuring machinesymbols.namesakeNon-deterministic Turing machinelawAlgorithm characterizationsPSPACEWolfram's 2-state 3-symbol Turing machineTuring machine examplesNSPACETuring reductionsymbolsUniversal Turing machineTime hierarchy theoremAlternating Turing machineRegister machine
researchProduct

Fuzzy Multi-Criteria Decision Making: An entropy-based approach to assess tourism sustainability

2019

In this article, we propose a method for ranking tourist destinations and evaluating their performances under a sustainability perspective: a fuzzy multiple criteria decision-making method is applied for determining sustainability performance values and ranking destinations accordingly. We select a set of sustainability evaluation criteria and use a fuzzy analytic hierarchy process to weight the selected criteria. We also optimize each evaluator’s membership function support by means of a fuzzy entropy maximization criteria. A case study is illustrated and results are compared with two data envelopment analysis–based models. The simplicity of the proposed approach along with the easy reada…

Tourism sustainabilitySettore SECS-S/06 - Metodi mat. dell'economia e Scienze Attuariali e FinanziarieOperations researchComputer science05 social sciencesGeography Planning and DevelopmentAnalytic hierarchy processFuzzy Multiple Criteria Decision-Making010501 environmental sciences01 natural sciencesFuzzy logicMulti criteria decisionTourismTourism Leisure and Hospitality Management0502 economics and businessSustainabilityanalytic hierarchy process DEA analysis entropy fuzzy logic multiple criteria decision-making tourism sustainabilityMultiple criteriaTourist destinationsEntropy (information theory)Fuzzy Multiple Criteria Decision-Making Fuzzy Analytic Hierarchy Process TourismSettore MAT/09 - Ricerca Operativa050212 sport leisure & tourism0105 earth and related environmental sciencesFuzzy Analytic Hierarchy Process
researchProduct

Five types of OECD healthcare systems: Empirical results of a deductive classification

2013

This article classifies 30 OECD healthcare systems according to a deductively generated typology by Rothgang and Wendt [1]. This typology distinguishes three core dimensions of the healthcare system: regulation, financing, and service provision, and three types of actors: state, societal, and private actors. We argue that there is a hierarchical relationship between the three dimensions, led by regulation, followed by financing and finally service provision, where the superior dimension restricts the nature of the subordinate dimensions. This hierarchy rule limits the number of theoretically plausible types to ten. To test our argument, we classify 30 OECD healthcare systems, mainly using O…

TypologyHierarchyActuarial sciencePublic economicsHealth Policymedia_common.quotation_subjectEmpirical ResearchTest (assessment)Core (game theory)State (polity)ArgumentSocial determinants of healthBusinessDimension (data warehouse)Delivery of Health CareOrganisation for Economic Co-Operation and Developmentmedia_commonHealth Policy
researchProduct

Kayne's model of Case and Finnish nominal phrases

2008

The standard view concerning Case assignment or valuation is that Case is valued to determiner phrases (DPs) in syntax. Recently, Kayne has proposed an alternative model, in which Case is valued to lexical elements rather than to phrases. This article cites several facts from Finnish in support of this model. A detailed Kaynean model of Case is developed. According to this model, abstract Case is valued to lexical elements by the highest ranking c-commanding Case assigner when each phase (CP,vP) is sealed, where ranking is based on a particular Case Hierarchy and a simple notion of locality. Configurations in which Case is seemingly assigned under a spec–head relation are provided with an a…

Valuation (logic)Linguistics and LanguageInterpretation (logic)Syntax (programming languages)Computer scienceLocalityDeterminerCase hierarchyLanguage and LinguisticsNominalizationLinguisticsRanking (information retrieval)Nordic Journal of Linguistics
researchProduct

A methodological system for hydraulic heritage assessment: a management tool

2016

Traditional irrigation systems are spatial units with a remarkable cultural and heritage value, which in turn generate water landscapes. They are part of hydraulic heritage, which includes material, ideational and symbolic assets. However, the assets related to historical irrigation systems and their landscape units have been poorly studied and evaluated. This paper develops a methodological system for the assessment of hydraulic heritage which combines basic and multi-criteria quantitative techniques. The authors conducted an evaluation of water heritage assets and their associated systems for the purpose of establishing a hierarchy for devising appropriate heritage management actions.

Value (ethics)EngineeringHierarchybusiness.industryEnvironmental resource managementWater supply010501 environmental sciences010502 geochemistry & geophysics01 natural sciencesManagement toolCultural heritage managementbusiness0105 earth and related environmental sciencesWater Science and TechnologyWater Supply
researchProduct