Back to Search
Start Over
Learning gradients via an early stopping gradient descent method
- Source :
-
Journal of Approximation Theory . Nov2010, Vol. 162 Issue 11, p1919-1944. 26p. - Publication Year :
- 2010
-
Abstract
- Abstract: We propose an early stopping algorithm for learning gradients. The motivation is to choose “useful” or “relevant” variables by a ranking method according to norms of partial derivatives in some function spaces. In the algorithm, we used an early stopping technique, instead of the classical Tikhonov regularization, to avoid over-fitting. After stating dimension-dependent learning rates valid for any dimension of the input space, we present a novel error bound when the dimension is large. Our novelty is the independence of power index of the learning rates on the dimension of the input space. [Copyright &y& Elsevier]
Details
- Language :
- English
- ISSN :
- 00219045
- Volume :
- 162
- Issue :
- 11
- Database :
- Academic Search Index
- Journal :
- Journal of Approximation Theory
- Publication Type :
- Academic Journal
- Accession number :
- 54650472
- Full Text :
- https://doi.org/10.1016/j.jat.2010.05.004