Royal Holloway logo and departmental theme Royal Holloway, University of London

The Nature of Statistical Learning Theory
Vladimir Vapnik, Springer, 1995

The Nature of Statistical Learning Theory cover This book discusses the fundamental ideas underlying the statistical theory of learning and generalization. The author concentrates on the main results and implications of learning theory, instead of proofs and technical details.

Written in a readable and concise style, the book is intended as an introduction to the subject of statistical learning for mathematicians, statisticians, physicists and computer scientists.

The subjects covered include:

  • setting learning problems based on the general model of minimising the risk function from empirical data;
  • comprehensive analysis of the empirical risk minimisation principle;
  • non-asymptotic bonds for the risk achieved using the empirical risk minimisation principle;
  • principles for controlling the the generalization ability of learning machines using small sample sizes based on these bounds;
  • a new type of universal learning machine to control the generalization ability.

Last updated Wed, 17-Jun-2009 15:57 GMT
Department of Computer Science, University of London, Egham, Surrey TW20 0EX
Tel/Fax : +44 (0)1784 443421 /439786