Statistical Mechanics of Learning: A Variational Approach for Real Data
- 19 August 2002
- journal article
- Published by American Physical Society (APS) in Physical Review Letters
- Vol. 89 (10) , 108302
- https://doi.org/10.1103/physrevlett.89.108302
Abstract
Using a variational technique, we generalize the statistical physics approach of learning from random examples to make it applicable to real data. We demonstrate the validity and relevance of our method by computing approximate estimators for generalization errors that are based on training data alone.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Statistical Physics of Spin Glasses and Information ProcessingPublished by Oxford University Press (OUP) ,2001
- Statistical Mechanics of LearningPublished by Cambridge University Press (CUP) ,2001
- Gaussian Processes for Classification: Mean-Field AlgorithmsNeural Computation, 2000
- Field Theories for Learning Probability DistributionsPhysical Review Letters, 1996
- Bayesian Learning for Neural NetworksPublished by Springer Nature ,1996
- Replica field theory for random manifoldsJournal de Physique I, 1991
- Spline Models for Observational DataPublished by Society for Industrial & Applied Mathematics (SIAM) ,1990
- The space of interactions in neural network modelsJournal of Physics A: General Physics, 1988