On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario
From MaRDI portal
Publication:3149529
DOI10.1162/089976602760128090zbMath1010.68115OpenAlexW1974755789WikidataQ74614436 ScholiaQ74614436MaRDI QIDQ3149529
Publication date: 25 September 2002
Published in: Neural Computation (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1162/089976602760128090
Related Items (13)
Search for a possible variation of the fine structure constant ⋮ Learning Coefficients of Layered Models When the True Distribution Mismatches the Singularities ⋮ Improving Generalization Performance of Natural Gradient Learning Using Optimized Regularization by NIC ⋮ On the Asymptotic Distribution of the Least-Squares Estimators in Unidentifiable Models ⋮ Relation between weight size and degree of over-fitting in neural network regression ⋮ Quantum information criteria for model selection in quantum state estimation ⋮ Parameter Identifiability in Statistical Machine Learning: A Review ⋮ Asymptotics for regression models under loss of identifiability ⋮ Tutorial on brain-inspired computing. II: Multilayer perceptron and natural gradient learning ⋮ Equations of states in singular statistical estimation ⋮ Asymptotic analysis of Bayesian generalization error with Newton diagram ⋮ Dynamics of Learning Near Singularities in Layered Networks ⋮ Variational Bayes Solution of Linear Neural Networks and Its Generalization Performance
Cites Work
- Stochastic complexity and modeling
- Estimating the dimension of a model
- Testing the order of a model using locally conic parametrization: Population mixtures and stationary ARMA processes
- Fitting autoregressive models for prediction
- Algebraic Analysis for Nonidentifiable Learning Machines
- Testing in locally conic models, and application to mixture models
This page was built for publication: On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario
