On the approximate realization of continuous mappings by neural networks
Explore this paper's citation graph
Summary
It is proved that any continuous mapping can be approximately realized by Rumelhart-Hinton-Williams' multilayer neural networks with at least one hidden layer whose output functions are sigmoid functions.
- Type
- article
- Published
- 1989-05-01
- Cited by
- 4,572
- References
- 24
- OpenAlex
- https://openalex.org/W1971735090
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:10203109
Keywords
Realization (probability), Artificial neural network, Computer science, Artificial intelligence, Mathematics
References
- A logical calculus of the ideas immanent in nervous activity
- Capabilities of three-layered perceptrons
- On the structure of continuous functions of several variables
- A Theory of Adaptive Pattern Classifiers
- Pattern Classification by Iteratively Determined Linear and Piecewise Linear Discriminant Functions
- Analog Perceptrons: On Additive Representation of Functions
- An introduction to computing with neural nets
- Noise reduction using connectionist models
- Phoneme recognition: neural networks vs. hidden Markov models vs. hidden Markov models
- Neural Net and Traditional Classifiers
- Phoneme Recognition: Neural Networks vs
- Physical and Biological Processing of Images
- Learning internal representations by error propagation
- Introduction to Computing
- Parallel Networks that Learn to Pronounce English Text
- Phoneme Recognition: Neural Networks vs. Hidden Markov Models
- An introduction to computing with neural nets
- Kolmogorov''s Mapping Neural Network Existence Theorem
- On the Representation of Continuous Functions of Several Variables as Superpositions of Continuous Functions of one Variable and Addition
Cited by
- Dynamic simulation of red mud washers used in aluminum industries
- PROCESS MONITORING OF ABRASIVE FLOW MACHINING USING A NEURAL NETWORK PREDICTIVE MODEL
- Performance Evaluation of Error Back Propagation Algorithm for Non-Linear Classificationand Function Approximation in VHDL Platform
- CSLS: CONNECTIONIST SYMBOLIC LEARNING SYSTEM
- Least Squares Solutions of the HJB Equation With Neural Network Value-Function Approximators
- Adaptive observers for unknown general nonlinear systems
- Existence and uniqueness results for neural network approximations
- Feedback-Linearization-Based Neural Adaptive Control for Unknown Nonaffine Nonlinear Discrete-Time Systems
- Blood pressure waveform characteristics exposed by spectral analysis and neural nets
- SYSTEM IDENTIFICATION USING DYNAMIC NEURAL NETWORKS: TRAINING AND INITIALIZATION ASPECTS
- Neural Networks Learning Differential Data
- On Adaptive Fuzzy Systems with Sinusoidal Membership Functions
- Efficient Approximation with Neural Networks: A Comparison of Gate Functions
- Internal model control using neural networks for nonlinear systems with delay
- On the Generalization Problem
- FUZZY NEURAL NETWORKS ARE UNIVERSAL APPROXIMATORS
- A study of prediction in seabed mapping
- Active data selection in supervised and unsupervised learning
- Dynamic planning model for determining cutting parameters using neural networks in feature-based process planning
- La estimación de la función de producción educativa en valor añadido mediante redes neuronales: una aplicación para el caso español
Related papers
- Primärzerlegung in Steinschen Algebren
- Über unirationale Scharen auf algebraischen Mannigfaltigkeiten
- Algebraic Set Operations, Multifunctions, and Indefinite Integrals
- On Almost α(Λ, sp)-continuous Multifunctions
- Produkttreue Klassen universeller Algebren
- Approximation of fixed points of multifunctions in partial metric spaces
- The World of Jagdish N. Srivastava
- Stratified realization of comparative probabilities
- Hardy spaces of analytic multifunctions.
- The equivalence of uninorms induced by the U-partial order