Learning Curves and Bootstrap Estimates for Inference with Gaussian Processes: A Statistical Mechanics Study 
Dörthe Malzahn, Manfred Opper

Abstract  We employ the replica method of statistical physics to study the average
case performance of learning systems. The new feature of our theory is
that general distributions of data can be treated, which enables
applications to real data. For a class of Bayesian prediction models
which are based on Gaussian processes, we discuss Bootstrap estimates
for learning curves. 
Keywords  Bootstrap, Gaussian processes, learning curves, statistical physics, variational methods 
Type  Journal paper [With referee] 
Journal  Complexity 
Editors  H. G. Schuster and K. Pawelzik 
Year  2003 Vol. 8 No. 4 pp. 5763 
Publisher  Wiley 
Electronic version(s)  [ps] 
BibTeX data  [bibtex] 
IMM Group(s)  Intelligent Signal Processing 