Learning from Examples as an Inverse Problem
The Journal of Machine Learning Research
Optimal Rates for the Regularized Least-Squares Algorithm
Foundations of Computational Mathematics
Dimensionality reduction and generalization
Proceedings of the 24th international conference on Machine learning
Spectral algorithms for supervised learning
Neural Computation
Elastic-net regularization in learning theory
Journal of Complexity
Thresholding projection estimators in functional linear models
Journal of Multivariate Analysis
On Learning with Integral Operators
The Journal of Machine Learning Research
Vector field learning via spectral filtering
ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part I
A practical use of regularization for supervised learning with kernel methods
Pattern Recognition Letters
| Hi-index | 0.03 |
In this paper we discuss a relation between Learning Theory and Regularization of linear ill-posed inverse problems. It is well known that Tikhonov regularization can be profitably used in the context of supervised learning, where it usually goes under the name of regularized least-squares algorithm. Moreover, the gradient descent algorithm was studied recently, which is an analog of Landweber regularization scheme. In this paper we show that a notion of regularization defined according to what is usually done for ill-posed inverse problems allows to derive learning algorithms which are consistent and provide a fast convergence rate. It turns out that for priors expressed in term of variable Hilbert scales in reproducing kernel Hilbert spaces our results for Tikhonov regularization match those in Smale and Zhou [Learning theory estimates via integral operators and their approximations, submitted for publication, retrievable at , 2005] and improve the results for Landweber iterations obtained in Yao et al. [On early stopping in gradient descent learning, Constructive Approximation (2005), submitted for publication]. The remarkable fact is that our analysis shows that the same properties are shared by a large class of learning algorithms which are essentially all the linear regularization schemes. The concept of operator monotone functions turns out to be an important tool for the analysis.