Explicit overall risk minimization transductive bound
暂无分享,去创建一个
Aside classical inductive methods transduction has reached an always increasing attention from the scientific community because of its learning paradigm. Explicit error bounds for inductive methods are well established results and stem from Vapnik theory or Rademacher complexity. In this work we address the problem of building an explicit form of the transductive bound presented in Vapnik Overall Risk Minimization approach.
[1] V. Vapnik. Estimation of Dependences Based on Empirical Data , 2006 .
[2] Bernhard Schölkopf,et al. A Discussion of Semi-Supervised Learning and Transduction , 2006, Semi-Supervised Learning.
[3] C. Scovel,et al. Concentration of the hypergeometric distribution , 2005 .
[4] Ran El-Yaniv,et al. Explicit Learning Curves for Transduction and Application to Clustering and Compression Algorithms , 2004, J. Artif. Intell. Res..