A note on the identifiability of the conditional expectation for the mixtures of neural networks

DOI10.1016/J.SPL.2007.09.038MaRDI QIDQ2483449zbMATH OpenOpenAlexFDO

Authors Jean-Pierre Stockis, Joseph Tadjuidje Kamgaing, Jürgen Franke

Publication date 28 April 2008

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.spl.2007.09.038

zbMATH Keywords

mixture modelsidentifiabilityneural networks

Mathematics Subject Classification ID

Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Neural nets and related approaches to inference from stochastic processes (62M45) Identification in stochastic control theory (93E12)

Recommendations

• Mixture of experts architectures for neural networks as a special case of conditional expectation formula.
• Mixture Models Based on Neural Network Averaging
• Modeling nonlinearities with mixtures-of-experts of time series models
• scientific article; zbMATH DE number 1304656
• Neural network identifiability for a family of sigmoidal nonlinearities

Cites work

• scientific article; zbMATH DE number 1324089 (Why is no real title available?)
• ARCH models and financial applications
• Estimating market risk with neural networks
• Local Global Neural Networks
• Prediction Intervals for Artificial Neural Networks
• Statistics of financial markets. An introduction.

Cited in

(6)

• Likelihood ratio of unidentifiable models and multilayer neural networks
• Mixtures of nonparametric autoregressions
• Neural network identifiability for a family of sigmoidal nonlinearities
• Mixture Models Based on Neural Network Averaging
• On geometric ergodicity of CHARME models
• Autoregressive processes with data-driven regime switching

This page was built for publication: A note on the identifiability of the conditional expectation for the mixtures of neural networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2483449)