Intrinsic dimension

Multivariate Statistics Neural Network Models

Published in Basilio de Bragança Pereira, Calyampudi Radhakrishna Rao, Fábio Borges de Oliveira, Statistical Learning Using Neural Networks, 2020

Basilio de Bragança Pereira, Calyampudi Radhakrishna Rao, Fábio Borges de Oliveira

A related concept in dimensional reduction methods is the intrinsic dimension of a sample, which is defined as the number of independent variables that explain satisfactorily the phenomenon. This is a loosely defined concept, and a trial and error process is usually used to obtain a satisfactory value for it.

A classification-based fuzzy-rules proxy model to assist in the full model selection problem in high volume datasets

View Article

Journal Information

Published in Journal of Experimental & Theoretical Artificial Intelligence, 2022

Angel Díaz-Pacheco, Carlos Alberto Reyes-Garcia

Another significant feature of datasets is its intrinsic dimension. The intrinsic dimension (ID) is the smallest number of parameters needed to represent data without information loss (Lombardi et al., 2011). The intrinsic dimension of the employed datasets was estimated with the (MNDE) and (DANCO) estimator (Ceruti et al., 2012). Estimation of the intrinsic dimension of each dataset is useful to ensure that each dataset represents a different computational problem. Based on this assumption, the proposed algorithms are capable of dealing with a wide range of problems and also with datasets of various sizes. In Table 4, the calculated intrinsic dimension using the estimators is shown.

Intrinsic dimension

Explore chapters and articles related to this topic

Multivariate Statistics Neural Network Models

A classification-based fuzzy-rules proxy model to assist in the full model selection problem in high volume datasets