Bayesian information criteria and smoothing parameter selection in radial basis function networks

Sadanori Konishi, Tomohiro Ando, Seiya Imoto

研究成果: Article査読

79 被引用数 (Scopus)

抄録

By extending Schwarz's (1978) basic idea we derive a Bayesian information criterion which enables us to evaluate models estimated by the maximum penalised likelihood method or the method of regularisation. The proposed criterion is applied to the choice of smoothing parameters and the number of basis functions in radial basis function network models. Monte Carlo experiments were conducted to examine the performance of the nonlinear modelling strategy of estimating the weight parameters by regularisation and then determining the adjusted parameters by the Bayesian information criterion. The simulation results show that our modelling procedure performs well in various situations.

本文言語English
ページ(範囲)27-43
ページ数17
ジャーナルBiometrika
91
1
DOI
出版ステータスPublished - 2004

ASJC Scopus subject areas

  • 統計学および確率
  • 数学 (全般)
  • 農業および生物科学(その他)
  • 農業および生物科学(全般)
  • 統計学、確率および不確実性
  • 応用数学

フィンガープリント

「Bayesian information criteria and smoothing parameter selection in radial basis function networks」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル