Generalization Error of Automatic Relevance Determination
Lists
Only you see this
Abstract
The automatic relevance determination (ARD) shows good performance in many applications. Recently, it has been applied to brain current estimation with the variational method. Although people who use the ARD tend to pay attention to one benefit of the ARD, sparsity, we, in this paper, focus on another benefit, generalization. In this paper, we clarify the generalization error of the ARD in the case that a class of prior distributions is used, and show that good generalization is caused by singularities of the ARD. Sparsity is not observed in that case, however, the mechanism that the singularities provide good generalization implies the mechanism that they also provide sparsity.
BibTeX
@inproceedings{nakajima2007generalization,
author = {Shinichi Nakajima and Sumio Watanabe},
title = {{Generalization Error of Automatic Relevance Determination}},
booktitle = {ICANN},
year = {2007},
doi = {10.1007/978-3-540-74690-4\_1},
}Loading related content...