ML Anthology

Test Error Fluctuations in Finite Linear Perceptrons

David Barber, David Saad, Peter Sollich
NeCo 1995 pp. 809-821
doi:10.1162/NECO.1995.7.4.809 /neco/1995/barber1995neco-test/

Abstract

We examine the fluctuations in the test error induced by random, finite, training and test sets for the linear perceptron of input dimension n with a spherically constrained weight vector. This variance enables us to address such issues as the partitioning of a data set into a test and training set. We find that the optimal assignment of the test set size scales with n2/3.

NeCo Semantic Scholar

Cite

Text

Barber et al. "Test Error Fluctuations in Finite Linear Perceptrons." Neural Computation, 1995. doi:10.1162/NECO.1995.7.4.809

Markdown

[Barber et al. "Test Error Fluctuations in Finite Linear Perceptrons." Neural Computation, 1995.](https://mlanthology.org/neco/1995/barber1995neco-test/) doi:10.1162/NECO.1995.7.4.809

BibTeX

@article{barber1995neco-test,
  title     = {{Test Error Fluctuations in Finite Linear Perceptrons}},
  author    = {Barber, David and Saad, David and Sollich, Peter},
  journal   = {Neural Computation},
  year      = {1995},
  pages     = {809-821},
  doi       = {10.1162/NECO.1995.7.4.809},
  volume    = {7},
  url       = {https://mlanthology.org/neco/1995/barber1995neco-test/}
}