Rigorous Bounds to Retarded Learning
Disordered Systems and Neural Networks
2009-11-07 v1
Abstract
We show that the lower bound to the critical fraction of data needed to infer (learn) the orientation of the anisotropy axis of a probability distribution, determined by Herschkowitz and Opper [Phys.Rev.Lett. 86, 2174 (2001)], is not always valid. If there is some structure in the data along the anisotropy axis, their analysis is incorrect, and learning is possible with much less data points.
Cite
@article{arxiv.cond-mat/0201256,
title = {Rigorous Bounds to Retarded Learning},
author = {Arnaud Buhot and Mirta B. Gordon and Jean-Pierre Nadal},
journal= {arXiv preprint arXiv:cond-mat/0201256},
year = {2009}
}
Comments
1 page, 1 figure. Comment accepted for publication in Physical Review Letters